top banner

work order tracking software

Operational Excellence in Manufacturing - delivering year-on-year performance improvement

Maintenance North West, conference, workshop, seminars and exhibition, Manchester (2009)

Conference Communication
Conference Communication has been trading for over 30 years and specialises in Industrial Maintenance Engineering and Asset Management, and organises Conferences and Workshops on these and related topics.

We publish Maintenance & Engineering (the bi-monthly magazine for maintenance professionals working in the Industrial, Commercial and Public Sectors). For twenty two years we published Maintenance & Asset Management (a subscription journal featuring more in-depth articles on maintenance management and related topics). From the January 2008 issue of Maintenance & Engineering the journal articles are incorporated into the magazine. John Harris, Editor of the journal for the past 23 years, is continuing to edit his articles. Maintenance & Engineering Magazine is the official media partner for easyFairs® MAINTEC 09 in the UK, with M&E producing the official MAINTEC Preview and the MAINTEC Exhibition Catalogue.

The combined magazine and journal are available in digital format and can be viewed by clicking on the links above. In addition, maintenanceonline.co.uk features the abstracts from every article published in the journal – copies of originals can be ordered from the site – and the site editorial pages feature articles of general interest to Maintenance and Engineering professionals.

We organised the first UK National Maintenance Engineering Conference in London in the early 1970’s and which, over the intervening years, evolved into the MAINTEC Exhibition at the National Exhibition Centre in Birmingham. MAINTEC, the leading Maintenance & Asset Management event in the UK was acquired by easyFairs® UK in March 2007 and, as easyFairs® MAINTEC, has been incorporated into a pan-european network of maintenance events.

We also distribute Maintenance, Manufacturing, Engineering and Management books drawn from various technical publishers; our Recruitment section features technical and managerial opportunities within industry and commerce; and using our Site Directory you can search our database of companies who offer products and services to either the Industrial Maintenance sector and/or the broader Factory Equipment/Services sector.

Simply click on the above links for further information. However, if you have a specific enquiry, either use the site search facility to access information contained within the site, or use the Maintenance Forum to put your question to a wider audience.

Driving Down The Cost and Environmental Impact of Road Haulage | The Business IMPACT of Enterprise Asset Management | The case for more comprehensive data collection and how it might be achieved: Part 1 | Applying PMBOK to Shutdowns, Turnarounds and Outages | Improved CMMS and Asset Management Systems - But do they lead to success? | Automated Trouble Shooting | Three into One does go | Too small for a CMMS? Think again | Equipment readiness and visibility using Honeycomb maps |

Driving Down The Cost and Environmental Impact of Road Haulage
One of the main trade associations for the commercial haulage industry, the Freight Transport Association, represents some 14,000 haulage companies who own in excess of 200,000 lorries – just 40% of the UK haulage fleet. In addition, its members operate over 1 million light vans. With the ever increasing cost of fuel and fuel duty FTA members and the freight industry as a whole can save considerable sums of money by optimising their fleet operations, not to mention the potential for significantly reducing the carbon footprint of the industry. Driving down the cost and environmental impact of road haulage Take the case of Dams International Ltd, one of the top four office furniture companies in the UK with approximately six per cent of the UK market share, who recently celebrated 40 years in business supplying not only the UK but the world with office and contract furniture. The company’s product range includes desks, storage, seating, screens and tables, and the company also provides furniture for small and home offices as well as for contract services. Dams International is currently saving over £750,000 per annum with the implementation of innovative transport optimisation software system which has enabled the company to radically improve the management of its drivers, vehicle maintenance staff and its 140 strong delivery fleet resulting in greater productivity and efficiency. The software functions include: routing and scheduling optimisation on a daily basis; managing transport resources; rationalising fixed routes; strategic planning and modelling; scheduling home deliveries continuously as orders are being confirmed; and managing the execution of the transport plan in real time using vehicle tracking technology. “We have adopted the Paragon Multi Depot fleet optimization system to deliver a range of operational benefits that are enabling us to make more deliveries and provide a better service with our available resources,” says Gary Grindlay, the Supply Chain Director for Dams International. He continues, “We have gone from five manual planning clerks who would take all day to plan delivery logistics to one Paragon planner who takes about an hour to produce our daily schedules. Before adopting the new system we made about eight deliveries per vehicle per day, whereas now we are doing 10 or more and we are aiming to extend this to about 12 each day.” “There are several reasons for these improvements. We can optimise vehicle loads by inputting weights and cubic measurements for each product, thus ensuring that vehicles are working to capacity. In addition we can plan appropriate delivery and installation times. This helps with Time and Motion management as we can quickly identify fitters who need additional training to bring them up to speed.” Dams International operates six sites to provide UK-wide coverage and the new system allows the company to assign deliveries from the nearest depot. This gives greater flexibility and ensures that vehicles and staff are kept busy. It also optimizes vehicle delivery mileage, keeping fuel costs down and reducing carbon emissions. “The system gives us much greater visibility via the built-in multi depot feature. The software automatically adjusts depot boundaries each day depending on live order volumes to ensure that our vehicles and fitters are fully occupied,” adds Gary Grindlay. “Using transport optimization software was the obvious choice for our operation as it gives us the planning reliability and business benefits we need to streamline our nationwide delivery operation. I have been in the logistics sector for more than 30 years and have seen the benefits such software can bring to other organisations. Paragon, in particular, is ahead of the game because the company specialises on transport planning and this shows in the quality of the software,” he concludes. Other companies benefiting from optimizing their transport schedules include Panasonic, John Lewis, B&Q, Tesco, Meile, Akzo Nobel, Marks & Spencer and Ikea, and Paragon claim that organisations can reduce transport costs by up to 20 per cent through more efficient deployment of vehicles and drivers, with the added bonus of reducing overall CO2 emissions. t: +44 (0)1306 732600 j.geary@paragonrouting.com
Click here to enquire..

The Business IMPACT of Enterprise Asset Management
The Business IMPACT of Enterprise Asset Management ‘Maintenance & Engineering’ is pleased to announce it is collaborating with IBM and David Berger of Western Management Consultants (the author of our regular ‘Transatlantic View’ column) on the serialisation of IBM’s new book ‘The Business Impact of Enterprise Asset Management’. The book runs to seven chapters and in this first instalment we feature the foreword written by David and Chapter One – ‘The changing role of the plant engineer’. A book of this nature allows you to zone in on specific measureable, realistic, and time-defined maintenance, repair and operation (MRO) goals that meet the demanding needs of your organization. Whether you are a VP of Operations, plant manager, facility supervisor, or shop floor professional, you reflect a basic industry trend: you are looking for data and a solution that can help you better align your specific activities with overall company goals and objectives. This book provides the guidance to meet many of these requirements by delivering a more holistic view of assets, along with the ability to execute according to specific asset management goals that are directly tied to operational performance. For further information, visit www.eamresourcecenter.com Foreword We all have assets, at home and in our businesses. We have always had assets, but never quite so seemingly complex. Technology has become an integral part of our lives, enabling us to get products and services better, faster, cheaper. That’s the good news. But if you’ve been on this planet for more than a few years, you know that there are no free rides – something has to give. I think you’ll find this book helps you sort out what the biggest trade-offs and challenges are in the world of asset management. As well, each chapter provides some practical solutions to consider, using Enterprise Asset Management software as an enabling tool. But most importantly, the book provides you with some answers to the “What’s in it for me?” question you will no doubt hear from key stakeholders across your company, as you try to get them excited about what Enterprise Asset Management can do for them, and vice versa. Like many of you, I enjoy a challenge. This book covers all the key ones that may have a significant effect on Enterprise Asset Management. For example, we are witnessing the growing complexity and risk associated with increased automation, as our need for better integration of technology increases. In terms of human assets, we have to figure out how to more effectively transfer knowledge from the heads of an ageing workforce into the hands of the next generation. One of the most difficult opportunities presented in this book, but potentially the most lucrative, is the challenge to grow the bottom line through lower energy costs, reduced emissions, and so on, as regulatory pressures rise and the cries to save our planet intensify. Roughly half of my career I’ve worked in industry, and half in consulting. One of the most satisfying accomplishments for me as an executive and a leader of people in industry, or as a consultant working with clients, is when the folks I work with experience an “Aha!”, that is to say, when they get it and their whole attitude and behaviour shifts dramatically. It’s when their face lights up and they are on a mission to make things better. So as you read through this book ask yourself two questions. First of all, what can you do differently to better manage your assets, today and over the long-term. Secondly, what can you do to motivate, influence and convince others to behave differently. That’s what leadership is all about. And that’s what this book is about. Enjoy it. Chapter 1 The changing role of the plant engineer Taking over the new world … If there is one lesson that IT has taught us in the last few years, it’s that nothing stands still. Not the world of business; not IT itself; and definitely not your job. Not that long ago, science fiction writers used to make up stories about the horror of robots taking over the world. But if they tried that now, they’d be laughed off the shelves. No one would be frightened – because we’ve all seen it happening. One recent survey1 showed that just over one third of all manufacturing plants in the US have at least 70 per cent automation and IT components on the plant floor. That figure is expected to increase to around 55 per cent of manufacturing plants in the next three years. Another interesting statistic in the study was that the maintenance of those IT-enabled assets is no longer just the job of the IT staff. The majority of plant managers said the job belonged to IT, and to the maintenance staff. So in the new world we’re in, managing the assets is no longer just for the IT specialist. The robots may be taking over the world – but we are taking over the robots. 1 Source: ‘Plant Engineering’ survey, November 2007 A whole new direction … For the plant engineer, it’s a challenging prospect. They’re used to change, of course – their role has always focused on making things, making them well. There’s been a clear trend over the last few years for anything that doesn’t contribute to the manufacturing process – equipment or facility maintenance or energy management, to name a few – to be outsourced to other departments, or even other companies. But today, managing IT-enabled equipment is a shared role between IT specialists and the plant engineers, and it demands people who have skills in both areas. The statistics have been gathered in the US, but industry observers are convinced that they represent a global trend – and that it’s a trend which will continue. As the wider world of business becomes less hierarchical, plant managers are increasingly involved in discussions about new construction and plant expansion, about production scheduling and financial management. Finance departments are looking to them for tighter budgets and new efficiencies; they are becoming involved in the development of purchasing policy in every field related to the manufacturing process, from compressed air systems to IT components and software. They are full partners in the business enterprise. It is this new relationship with the IT department that is most significant. In an area that was once the jealously-guarded preserve of the specialist, plant engi- neers are being called on to use their experience and knowledge of how the plant works to provide important insight in drawing up the specifications for new Enterprise Asset Management (EAM) solutions. Plant managers always had experience and expertise to offer, but now the vast majority of them have degrees – and nearly a quarter have an advanced degree as well. There is a whole new layer to their responsibility within the organisation: now they are seen as being responsible for cutting costs, improving performance, and driving efficiency. New challenges … That is good news both for the plant managers and for their companies – latest figures show their 2007 bonuses up by nearly 46 per cent from 2006, the second time in two years they have shown sharp increases. Obviously, that’s good for the plant managers – and good for the companies because it shows that their productivity goals are being met. But it’s a competitive world, and a chilly economic climate. It’s not enough to improve over one year. You have to keep doing it. And that’s where the picture gets a little gloomier. It’s not only the pwlant managers who have new roles and new responsibilities – their staff, too, have to adapt to the changing global business world. There are also new challenges to face throughout the company, as responsibility spreads downwards. The trend in the manufacturing sector is towards sharing responsibility for predictive maintenance of assets between full-time maintenance staff and the operators of the machines in question. That’s the logic of increased IT-enabled equipment. But at present, many firms – 40 per cent or more, according to one study1 – don’t even have a predictive maintenance program in development, let alone up and running. And what’s more, the IT functionality of many assets simply isn’t used to its full potential. The plant manager’s own IT competence isn’t enough. To get the real benefits of the technology, he needs to be able to improve the level of training among the operators of the equipment and throughout the organisation. That’s where the extra cost savings and the efficiency improvements, that he is being asked to produce, are to be found. The technological solution … Technology isn’t the problem – it’s the answer. The pace of competition means that assets are being run more aggressively, with less room for forgiveness, and in an unpredictable regulatory environment. Increased automation throughout the asset management process has the potential to free the plant engineer and his whole department. New developments such as wireless sensors and activators – “smart dust” technology – will eventually load sensors, power supply, Circuitry, and microprocessors onto a mote the size of a speck of dust, and the growth of broadband and VoIP technology will revolutionise management and maintenance of sophisticated manufacturing equipment. Handheld devices, rugged PCs, automated data entry, and satellite tracking systems mean that both fixed and mobile assets can be monitored, tracked, and maintained from a single computer platform. These technologies aren’t only being used in the manufacturing of hard goods either. In pharmaceuticals, petrochemicals, food and retail industries, these technologies are finding acceptance. They are driving productivity gains at every stage of the enterprise, from procurement of raw materials to delivery of finished goods. Taking control of the revolution … Globalisation; new technology; soaring energy prices – the pressures on global manufacturers is intense, and the result has been nothing short of a revolution. It’s almost impossible to remember what life was like in the simpler days of only a few years ago. But any historian knows that it is impossible to take control of a revolution in full flood. This time, with the available technology and the experienced plant engineers to drive it, it might just be different. 1. Source: 'Plant Engineering' survey, November 2007
Click here to enquire..

The case for more comprehensive data collection and how it might be achieved: Part 1	Conference Communication
The case for more comprehensive data collection and how it might be achieved: Part 1 David Sherwin Former Professor of Terotechnology, Lund and Växjö Universities, Sweden, now retired Abstract This paper is about possible methods for the practical application of Terotechnology. It is explained that data analysis is prerequisite to formulating, and later updating, optimal maintenance and plant renewal policies. The nature of optimal maintenance policies in the context of the Life Cycle Profit (LCP) method of Hans Ahlmann is discussed, as are the requirements for applying this method, the most important of which are (i) a better database than most managers of companies are used to, and (ii) better software to analyse the data. It is explained that, for productive systems, the long-term costs involved in data collection to permit full realisation of LCP are unlikely to exceed the long-term benefits of a management policy based upon Total Quality Manufacturing and LCP rather than Management by Objectives and Reliability Centred Maintenance, this being particularly so if data requirements for the firm are integrated. In this first part of the paper, the author explains why he believes that management science has failed to integrate or deal optimally with the maintenance function and discusses the need for, and availability of, statistical modelling aimed at the optimisation of maintenance policy. In the second and concluding part of the paper, to be published in the next issue, he will offer a detailed critique of combined systems of overall management and maintenance optimisation and explain why he believes that there is a strong case for comprehensive data collection. Keywords: LCP, LCC, Database, IT, Integration of business functions, Terotechnology INTRODUCTION AND OVERVIEW General This paper is a considered, but essentially personal and opinionated, view of how the subject should be handled in practical applications for the best advantage of all the stakeholders in a productive undertaking. This view is based on a total of forty years activity in the general field of maintenance, reliability and quality, only half of which was in universities. Maintenance and replacement are serious matters, if only because they together account for 10-40% of total costs in an enterprise. Their effects on other functions such as Production and Quality are often multiplicative, yet they seldom appear as a vital factor on the organograms for schemes for Computer Integrated Manufacture (CIM) or Supply Chain Management (SCM). Three strands have been discernible in both academic and practical thinking and action about maintenance over the last few decades. The Management Science (MS) approach treats maintenance as a cost. The Operational Research (OR) approach treats it as a problem in mathematical optimisation. The third approach, typified by Reliability Centred Maintenance (RCM) and Total Productive Maintenance (TPM), consists of simple – but inevitably flawed or incomplete – models of reality which attempt to reconcile the other two approaches with the experience of practical maintenance engineers, and the other needs of the maintenance function. All three approaches suffer from the failure of their advocates to take account of one or more important aspects of the overall problem. The three strands really should be laid up into a rope, in which each is supported by the other two and the whole is stronger than the sum of the parts. This paper argues the case for a more comprehensive approach, based on Life-cycle Profit (LCP) and Total Quality Manufacturing (TQM). In this more holistic terotechnological, approach, maintenance is no longer treated as an isolated function, but integrated with the other functions of the firm, and justified by estimating and later measuring its contribution to overall profitability. Such an integrated approach requires the support of an inclusive IT system with potential benefits far beyond the maintenance function and this year’s bottom line. The Management Science approach This suffers from advocates, managers and theorists who do not understand technology. Drucker’s ‘Management by Objectives’¹ is typical. Management scientists persisted with the belief that maintenance is a fixed cost that is only reducible when hard times reduce the requirement to use the machinery to be maintained. Only very recently has anything appeared in the MS press that acknowledges the connections between maintenance and quality and market share, and even then there is no attempt to find useable quantitative methods. The Life-cycle Cost (LCC) and later, Life Cycle Profit (LCP), approaches have come from engineering rather than the management academics, and so have not found ready acceptance by managements of large concerns outside Scandinavia. On the MS side, there have been such movements as Kaplan and Norton’s Balanced Scorecard ² which acknowledge the importance of factors other than the bottom line, but still keep them in watertight compartments. Other fads, invented or taken up by management scientists, concentrate upon one aspect or one technique, often over-simplified for easy popularity, e.g. Six-Sigma³. The crazes for such fads seldom last more than ten years, which is just as well because most of them turn out to be ultimately harmful, even though they are usually not blamed directly by management coroner-pathologists or bankruptcy receivers. This is because of the relatively long time between execution and effect in such cases. (For crazes from within Maintenance/Terotechnology itself see later in the section on Combined Systems of Management and Optimisation) The mathematical modelling approach A plethora of mostly inappropriate mathematical models has poured out of academic departments of mathematics and operational research (OR) for many years. Surveys ^4,5 have shown that very few, 20% say, of these models are potentially useful, and even fewer, about 1%, have been reported as actually used. Even then, they are often used inappropriately. The principal problem has been misunderstanding of the reliability theory appropriate to maintained systems, the so-called ‘bathtub curve’ problem^6,7. This applies to the consultants and OR specialists as well as their (potential) customers, the engineers managing the maintenance function. The principal generalised faults of the OR specialists and applied mathematicians have been to be satisfied too easily with inadequate data, and to draw dubious conclusions from over-elaborate and mathematically difficult models that attempt to make the best of what can only be, in the end, a bad job due to inappropriate data. The general statistical principle that the higher the information content the better the best possible decision can be seems to have been lost, together with the original concept of OR, defined as the application of scientific method to operational problems. Engineers will not apply models they do not understand, especially when individual results challenge their common sense. As a result, few of the models devised by operational researchers and mathematicians have been applied, other than simple Age and Block Renewal, (AR, BR)⁸, and even these are often mis-applied to complex systems in which only the failed parts are renewed; bad examples⁹ were criticised at the time of their publication, and later⁷. Models developed by mathematicians for the case where faults develop and are mostly detected by inspection before they cause failure have usually been too complicated for general application and too inaccurate to save serious money, with the notable exception of those of the Salford school¹⁰. Combined systems of management and optimisation The general faults described above, and many more, are included in RCM¹¹, currently one of the most popular maintenance management models. For a full exposé of the problems with RCM see the author’s own paper¹². TPM provides a more useful way of connecting maintenance to quality and plant economics than Six-Sigma, but nevertheless has limitations, only partly and very imperfectly covered by RCM¹³. All three approaches are somewhat faddish because – (a) they are limited in their scope and based, explicitly or implicitly, upon incomplete or even false models of reality, and/or – (b) they play on the manager’s natural desire for simple prescriptive solutions to complex problems that are not amenable to such solutions, and/or – (c) they are promoted and oversold by consultants who themselves do not fully understand the problems, and/or – (d) they contain false measures of success that ‘demonstrate’ that they have been effective. (A fad is defined in the Concise Oxford Dictionary as “a craze, a fashion… a piece of fancied enlightenment” and that is also the author’s meaning). Solutions - the need for data collection and functional integration After examining the failures and inadequacies of current practice, expanding under each of the headings above, some solutions will be suggested. These all involve more detailed and more complete data collection, to make possible the more sophisticated analysis and functional and mathematical modelling needed for real and continuous improvement. However, the second and more important purpose of this paper is to discuss the need for, and benefits of, a more integrated approach that takes account of the interactions of the traditional functions or departmental responsibilities in a productive enterprise. We think that just ‘considering’ these functions in a bold diagram with lots of arrows and circles is not sufficient; to get ahead of the field and stay there, it is necessary to know – or at least be able to estimate and continuously refine – the effects that changes in one function have upon the others. In this regard, Maintenance is but one of many functions with a levered effect far beyond its own internal costs and concerns. MANAGEMENT SCIENCE AND MAINTENANCE In this section we examine in more detail the assertions made above regarding the failure of management science to integrate or deal optimally with the maintenance function. The changes necessary in the training and practice of general managers and engineers to improve the situation will also be discussed. Managerial misconceptions Almost all of the standard managerial and production economics texts treat maintenance as an expense or even as a fixed cost. They regard it as unavoidable, and necessary to make all their other assumptions about the failure and performance of the machinery true. According to most of these texts, the machine’s maker’s instructions are to be followed without question or variation. No account is taken of the need for more or less maintenance depending on the severity and intensity of use, and no changes to the schedule are permitted in the light of experience, except to ‘save money’ on the maintenance budget. Even books on advanced manufacturing management techniques, such as Computer Integrated Manufacturing (CIM), have no box for maintenance on their organograms and flowcharts. Production plans go awry because the inevitability of stoppages for adjustments and failures has been ignored. This problem has become more acute because of the popularity of Just-in-Time (JIT) manufacturing methodology and ruthless reductions in stocks of raw materials, work-in-progress and finished goods. The MBA-course view of maintenance extends to other functions of a technical nature such as design and production. All three are usually to be run to rigid rules, which the non-technically-trained manager dare not change. Instead, they concentrate on finance, sales and marketing, which they presumably do understand, and so companies fail for lack of technical feedback and innovation through research and development. Engineers are sometimes equally guilty. A manufacturer of small machined brass castings complained to the author of falling revenues and sales. When questioned as to the quality of his products, and the modernity of his machinery and methods, he rejected any idea that they were inadequate to the present-day requirements of his customers. Yet most of his machinery was more than thirty years old and 30% of product, mainly pressure relief valves for domestic boilers, was rejectable at first inspection. To sustain their precision, his antiquated machines needed more maintenance than they got, and far too many castings were porous, due to inadequate temperature control of the molten metal. These drawbacks were accepted because they had either always been there or had built up very slowly. The only reason that any quality or maintenance records were kept was to satisfy the requirements of the British Standard for relief valves. Like most standards, this one was not concerned with the manufacturer’s economics, only the quality and safety of the product reaching the user. As a result, the company was beaten for price for their main product by a competitor for whom boiler relief valves were a sideline Our advice, to modernise, diversify and collect data to guide further advances, was ignored. The factory, one of the oldest in Birmingham, was converted to a block of yuppie flats with a night club in the basement. This is an extreme example, but by no means unique. An almost parallel case occurred in Queensland, the only difference being the products, which in this latter case were domestic water heaters and tanks. In both cases, sons with Business Studies degrees had succeeded engineer founding fathers. Management by objectives (MBO) When he conceived MBO¹ Drucker, can have had no idea how it would be distorted and over-simplified by consultants. As usually applied, MBO inevitably, if unintentionally, encourages managers to meet their own targets without regard to the effects on others or the company as a whole. Senior managers are advised to write down their overall aims and then to subdivide them into targets for their immediate juniors, and so on down the line. It has been referred to as ‘silo management’ with the various managers each in their own silos, unable to see or care what is happening in any of the other departments. Consider a simple example that could occur in any manufacturing company. At the start of the year, the general manager discusses separately with the managers of Maintenance, Production, Quality, Sales and Stores what their targets should be for the next twelve months. The Maintenance Manager reluctantly agrees to cut his budget by (another!) 10% and is promised (verbally of course) that the board will approve new machinery next year, provided that these and other savings are achieved. The Production Manager is asked to increase production by 10% and agrees despite misgivings about the Sales department’s ability to sell the extra goods. He dare not say so but is planning to achieve the new target by cutting a few corners on quality. Unbeknown to him, however, the Quality Manager has agreed to reduce the number of customer complaints by 50%. The Stores Manager agrees to a 10% cut in inventory and she secretly intends to sell off any spare parts that have not shifted in the last two years. The Sales Manager only agrees to try to sell the extra 10% on the understanding that quality will improve. Clearly, these managers will soon be at each other’s throats, and some of them will be fired for missing their incompatible targets. Under the prevailing downsizing aegis, their jobs are doubled up among those frightened and overworked managers who remain, and further decline takes place despite their agreement to new targets. This is sold to the shareholders as necessary adjustment to prevailing market conditions, for which all board members of course deserve a fat bonus. Consultants and core competencies In past papers, the author and others have blamed such vicious cycles as the above on failure to address quality problems and the need to refresh the product cycle, but it is now clear that a more general theory, based on ‘joined up’ thinking and inter-departmental co-operation is needed for the increasingly competitive future. But first, the managers must stop competing with each other and start to co-operate to achieve reasonable corporate targets. They will not do this unless they feel secure. Their underlings will not co-operate unless they also see some end to the downsizing and despair. This will not occur simply because all the managers retire to the countryside for a week of pointless exercises carried out in intense physical discomfort. Nor should they blindly follow the advice of consultants. Consultants usually feel that their reputation depends upon quick rather than lasting improvements, which leads to the curse of financial short-termism, pandemic in the English-speaking world. Historical research by American economists, no less, has revealed that flexibility, quality and innovation rather than retreat to so-called ‘core competencies’ are characteristic of long-lived companies with good labour and customer relations¹⁴. These economists failed to acknowledge the considerable body of Total Quality Management and terotechnological literature, because they would never even consider reading anything written by an engineer. The Swedish company Stora (which means ‘Big’) dates (at a bit of a stretch) from the 13th century and has only recently succumbed to a takeover. It started as a timber concern, but rather than abandon its development plans, moved into hydro-electric power, because it was needed if they were to develop their paper industry and became a major player. Nokia used to make rubber boots. Another example is Siemens, which has grown by always trying to be first with the best. Thompson in France, Phillips in Holland, GE in the USA, ABB in Scandinavia and Switzerland, GEC in Britain, the big Japanese combines and BHP in Australia all grew and prospered by diversification and innovation and all have suffered setbacks (in GEC’s case almost fatal) when they tried to concentrate upon what they thought they knew best. Just as the departments in a single company should co-operate selflessly to endure, the individual companies in a conglomerate should all aim to maximise the stability and profitability of the whole rather than the divisions separately. New products can be developed only because others are profitable, and must be developed because the present ones will not remain profitable indefinitely. Yet in engineering companies world-wide, businesses built up over a century or more by the instinctive application of these principles have been destroyed in a decade by combinations of overpaid boardroom incompetents and ignorant self-styled consultants applying distorted management theories advising a return to ‘core competencies’. They reduce the company until even they can manage it, shedding the green offshoots of future growth. Australia has failed to grasp the opportunities offered by plentiful indigenous raw materials and an isolated home market. The raw materials are exported and come back as high value-added products. Wool is exported to return as clothing with Italian and British labels. Chrome and nickel ores go all the way from Western Australia to Finland, to return as stainless steel products. All the raw materials for the manufacture of aircraft and jet engines exist in Queensland and the climate west of the dividing mountain range is perfect for their assembly. Boeing is in Seattle because that is where the spruce was found; they spend a fortune on correcting the wet atmosphere so they can use modern adhesives. The Australian wine industry demonstrates that it need not be this way, although perhaps it now stands in danger of ‘over-chardonnisation’. This under-investment culture was imported from the UK. The education of managers of technological enterprises Business Studies used to be a purely post-graduate, post-experience, operation and probably should be still. Some of its components, such as accountancy and marketing, were taught separately at undergraduate level, but the undergraduate degree that combined most of them was still based on, and called, Economics, and was sometimes combined with Mechanical or Production Engineering. Graduates of such courses had some hope of eventually becoming competent to run businesses in the financial or engineering sectors respectively, particularly if they invested, preferably after some practical junior experience, in an MBA course. In contrast, the modern BS graduate knows nothing about technology but expects to be telling experienced engineers what to do just a few years after graduation. Because governments make company law and taxation so complicated many companies in the English-speaking world think that these bean counters must be in charge to keep the company out of the courts. Actually, many examples, present and historical, show that engineers and scientists are quite capable of running a business, given the appropriate training. In fact, they usually do it better than the accountants and BS graduates because of their generally higher mathematical ability. Conclusion Management Science has contributed to industrial growth in the past and many of its methods remain valid, but its practitioners are failing to optimise industrial efforts world-wide, mainly because they have abandoned scientific method in favour of simplistic fads and short-termism, but also because technological aspects have been under-emphasised. We shall examine some of these fads in Part II of this paper, after we have looked at proven tools from both Management Science and Terotechnology, and will suggest ways to use them in combination to multiply their effects and so stabilise and expand a productive organisation. THE PLACE OF MATHEMATICAL MODELLING Introduction During World War II aircraft maintenance was one of the first areas to be tackled by the early operational researchers . They correctly distinguished between war and peacetime requirements and prevented the withdrawal for maintenance of serviceable aircraft that would probably be shot down before anything vital failed. The peacetime schedules, quite properly designed to sustain a high level of readiness (for war), were ignored in favour of damage repair and a few routine checks and oil changes. This is a totally different problem from that posed by complex manufacturing systems. But whatever the context, we may infer that some form of statistical model of the incidence of failure, the time to repair (and the effect on both of the stress levels on the machinery) and of the costs and benefits to the company as a whole, is necessary to the optimisation of a system’s maintenance policy. The model must reflect reality sufficiently accurately to avoid ‘Garbage-in Garbage-out’ (GIG), which can usually only be warranted by an engineer and is beyond the capabilities of BS graduates with no technical training or experience. Unfortunately, few engineers have the mathematical knowledge to analyse the data, infer a mathematical or statistical model and calculate the relevant optima. It has taken the author over twenty years to acquire sufficient insight and mathematical knowledge to be reasonably confident. Inappropriate mathematical models can be startlingly counter-productive. With these factors in mind, we examine now the nature, classification and use of mathematical models, bearing in mind that there are many more published models than there are recorded applications. Classification of models Models may be dichotomously classified in four ways, giving eight possible combinations, with respect to their purpose (see Figure 1). The most important distinction is between models for components and models for systems. We should never forget that systems fail but we repair parts. System models must therefore be based upon analysis of data relating to parts. Maintenance schedules should not only be optimised as to frequency but must be specific about what is to be maintained and how. This obvious statement is, in practice, too often ignored. Figure 1. Taxonomy of maintenance models Stochastic models are those in which we determine, from the statistics of failures, the interval between interventions at which the expected costs (benefits) of lifetime maintenance and replacement are minimised. In a deterministic model, interventions are determined by a physical change, or occur when some variable other than age crosses a pre-determined limit. Falling between these two are inspection models in which the intervals are statistical but the decisions are deterministic. Models can be based on the assumption either of a finite or an infinite system life. The latter case effectively approximates that in which the lives of fallible parts are only a small fraction of the expected endurance of the system as a whole. Similarly, some models are founded upon the statistics relating to the variability of failure and repair times, while others call for maintenance on the basis of a physical measurement. Finally, it is vital that we distinguish between systems whose parts are renewed and the behaviour of the parts themselves, because different modelling techniques are needed. System models should usually be built up from models of the behaviour of the fallible parts within the system. Maintenance modelling of productive systems Although it is convenient to develop the theory from consideration of systems that are used to manufacture goods, its applicability is in fact wider because all systems have a purpose, which may be considered to be a product. A long series system whose modules are serially preventively maintained by a workforce that cannot tackle more than one (or very few) thing(s) simultaneously inevitably spends most of its life well maintained but unserviceable. The economics of system maintenance have therefore come to depend more on the purpose than the physical nature of the system, and availability is now seen to be as important as speed and technical capability (accuracy and precision). In particular, the true cost of downtime is seldom calculated accurately, which results in under-manned maintenance squads and delays while contractors are brought to site to deal with failures. This is why many manufacturing systems still rely upon buffer stores between stages to cover failures. Just-in-Time (JIT) is really just a crude psychological trick to make everyone more careful to avoid breakdowns; attempts at applying pure JIT have often resulted in either re-introduction of buffer stores or duplication of machines. It is a common error to suppose that reliability is more important in manufacturing systems than availability, and another to insist that the system availability in a series is the product of the part availabilities. The first error arises from the military background associated with reliability theory. In war, reliability over a mission may well be important, but availability to start the mission is obviously rather more so. In manufacturing, the emphasis changes because operation is ideally more continuous, but reliability may still be a factor in prompt delivery or avoidance of start-up losses. The second error arises from failure to consider that a system that is shut down due to failure of a part is not at risk of further failures until it is restarted after repair. This is not important to accuracy in short series with some parts of relatively low availability, but in long series and where all the part availabilities are of the same order, it is vital for the acceptability of a proposed system. The correct calculation is given in Equation 1 as follows: This may be proved using nothing more complicated than a Venn diagram (see Sherwin and Bossche [15]), yet very few textbooks get it right and none except our own acknowledges its significance, which can be demonstrated by considering a series system of, say, four hundred parts each with Availability a = 0.999. We considered in that same book the productiveness of systems with some redundancy, productiveness being defined as the actual possible long-term average output rate expressed as a fraction of that with perfect reliability, i.e. no failures. It therefore depends significantly on the throughput capacity of the least productive stage of the system. Productiveness differs from availability in that the possibility of production at lower rates during the failure times of partially redundant machines in the system is taken into account when calculating the long-term system mean output. Only in a straight series system are availability and productiveness interchangeable. Even the revised availability calculation will not do for productiveness: each possible state of the system must be considered, the system productiveness being the sum of the products of the stage-state probabilities and their corresponding output rates. Many manufacturing systems enter service without such calculations having been made, with the result that they do not perform adequately when stretched by a successful product. Hurry to fulfil orders then leads to acceptance of sub-quality product and spares, botched repairs and neglect of necessary preventive maintenance. The items or stages in a series system are themselves complex. They each consist of parts, some of which benefit from preventive maintenance (PM). *Systems fail but we repair or renew parts.* It follows that data collection analysis and optimisation of intervals should be at the parts level, initially. The times between failures for a machine or system have, however, often been assiduously collected and analysed – but disregarding which parts have failed⁷. In some cases, PM has been discontinued because analysis fits a Poisson pattern. It is true that a maintained system with or without PM often has a sensibly constant Rate of Occurrence of Failures (ROCOF), but it is also true that ROCOF can be reduced by PM, and that there exists an ideal PM + inspection schedule that minimises the combined downtime or the total cost, or maximises the long-term expected profit. This schedule is a function of the statistical failure characteristics of the fallible parts (noting that over 80% of engineering parts either outlast (or determine) the system’s useful life or else are best left to fail). We also deplore the careless habit of lumping downtime due to PM (which can often be done in parallel and/or without loss of planned output) with that due to failures, which is stochastic but partly dependent on the frequency and quality of the PM. Review of models for optimising parts maintenance There are two basic kinds of model for parts, based respectively upon age since last renewal and upon measurement of some indicative variable (see, for example, Jardine,⁸). The variations on the first theme are well known. The cost of failure must exceed that of PM, including downtime costs in both cases, and the expected number of failures divided by the time since last renewal must be increasing. The models are – (a) Age Renewal, in which the part is run until it either fails or reaches an optimum age t, (b) Block Renewal, in which renewals occur to a fixed schedule and failures are renewed or repaired as they may occur, and – (c) Bad-as-Old Renewal, which is Block Renewal with the difference that failures are restored only to the pre-failure condition. None of these ever represents reality exactly, but each is useful as a fairly close approximation or as a limiting value in different circumstances. Given the costs and an estimate of the underlying distribution of failure times in the absence of PM, optima can be estimated. All except the Bad-as-Old model require some tedious calculations to find the optima, but in the computer age this is not really a problem. The part is considered in isolation from the rest of the system, i.e. the models do not consider any relationships that there may be with other parts. In practice it may be so difficult to get at a part that it becomes economic to consider renewing other unfailed parts at the same time. In other cases, it can pay to perform a bunch of routines at the same stoppage in order to reduce expensive downtime, with none being done at their individual optima. Often, the original equipment manufacturer (OEM) guesses at the intervals required and issues a recommended schedule that is never challenged by the operator because no data are collected and the OEM will not pay out on the warranty if the schedule is not followed. But these costs are negative benefits, and needed to confirm project viability and to assess profits. The second type of model requires the operator to monitor, continuously or at fixed or calculated intervals, some variables that allow him to judge the part’s condition. By recording and graphing the readings, the item can be taken close to failure before being maintained. Failure is either graceful decline to a defined unsatisfactory limit or else is preceded by a sudden change in the level or gradient of the graph. Provided that monitoring costs are not too high and that indication of imminent failure is accurate, inspection models improve availability and productiveness. Recently, the cost of monitoring has been declining and more cases can be justified, but it is still being applied without making proper predictions of the cost savings or full assessment of the accuracy. Frequently, no model is made at all and unsubstantiated claims are made concerning savings. However, it is true to say that in cases in which the downtime costs far outweigh the material and labour costs, inspection or monitoring will be a strong contender, provided all the facts are known and properly modelled^{16, 17}. Premature removal is a big problem in inspection or monitoring; there is a loss of faith if the removed part shows no wear or damage. Review of models for systems The useful system models fall into three categories. First, there are many variations on the theme of combined Block models. The second category aims to optimise overhaul intervals either by tracking the rise in ROCOF since the previous overhaul or by inspection or a combination of both. The third category, Opportunity Maintenance, is possibly the most suitable for systems that must operate continuously, and can be modified for systems with scheduled shutdowns¹⁸. The Combined Block Model shapes the traditional schedule of grouped actions at multiples of a common interval. The common interval should be chosen so that downtime or cost due to failures and PM is minimised over the system, but in practice it is seldom optimised. The basic interval is usually fixed by statutory, logistic or other considerations and the best schedule found using its multiples. For example, a factory boiler may be cleaned and maintained over a weekend when its output is not needed; optimisation then consists in finding the best number of weeks. Heavier work would be left to the summer when the boiler is not needed at all. Although no item is done at its independent optimum, time and money are saved by the grouping. Provided that legal and contractual difficulties can be overcome, a properly calculated Combined Block schedule can be economic, particularly if the workforce can find other work between periods of maintenance of the system in question. It can be fairly easily modified to accommodate some items that are monitored or inspected, particularly if the inspections require the system to be stopped. Reliability theory explains how a complex system that is maintained only at failure and then only to repair the failure, settles to a constant average ROCOF⁶. If, in addition, the system gets PM at relatively short intervals, then the residual ROCOF is reduced because some failures are prevented. As the intervals are lengthened, the average ROCOF rises and vice versa. If some of the routines are concentrated into overhauls, then the ROCOF remains constant on average but rises from a low point after each overhaul. (There may be a temporary rise after each overhaul, but this is short-lived and due to faulty work and poor quality spares.) It can be shown that this rise in ROCOF is theoretically exponential; that is, if the overhaul is sufficiently delayed it will level out to a higher constant value, again on average. If the corresponding cost (or better still, nett benefit) curve can be traced, then the optimum overhaul interval, given a pre-determined schedule of more minor PM, can be found. Alternatively, overhaul can be made partly or wholly dependent upon inspected or monitored condition. In the latter case, there is usually a single vital item that usually or always determines when the overhaul is done. The problem remains as to which items to maintain outside of the overhauls; some will be obvious but others marginal and the calculations are possible but not easy. A fuller discussion of Opportunity Maintenance and how it might be op timised has been given by the author¹⁸. The basic principle is that it may well be better to wait until something fails and then take the opportunity to perform other PM that is nearly or over-due during the enforced stoppage. A preliminary model for a continuously required system was given. Variations include different rules as to how long an enforced stoppage can be extended for PM, and for stopping anyway if a failure does not occur naturally within a certain interval. All the system models discussed above require as inputs complete parts data with respect to failure time distributions, repair and PM times and costs. All depend upon first calculating policies for parts, which are modified in the full system models. REFERENCES* 1. Drucker P F, The Practice Of Management, Pan Books, 1968 2. Kaplan R S and Norton D P, The Balanced Scorecard, Harvard Business School Press 1996 3. Magnusson, Kroslid and Bergman, Six Sigma: the Pragmatic Approach, Studentlitteratur, Lund 2000 4. Valdez-Flores C and Feldman R M, A survey of preventive maintenance models for stochastically deteriorating single-unit systems, Naval Research Logistics, Vol 36, pp 419-46, 1989 5. Dekker R, Applications of maintenance optimisation models, Report No.9228/A, Tinbergen Econometric Institute, Erasmus University, Rotterdam, 1992 6. Ascher H E and Feingold H, Repairable Systems Reliability, Modelling, Inference, Misconceptions and their Causes, Marcel Dekker, New York and Basel, 1984 7. Sherwin D J and Ascher H E, Reliability data analysis for economic advantage: problems concerning time windows and repairable systems, IFRIM Conference, Växjö, Sweden, 2002 (available by e-mail on application to the author) 8. Jardine A K S, Maintenance, Replacement and Reliability, Pitman, London, 1973 9. Ansell J and Phillips M J, Practical Methods for Reliability Data Analysis, Oxford: Clarendon Press, 1994 10. Christer A H and Waller W M, Delay time models of industrial inspection maintenance, Journal of the Operational Research Society, Vol 35, pp 401-6, 1984 11. Moubray J, Reliability-Centred Maintenance, Butterworth-Heinemann, London 1991 12. Sherwin D J, A critical analysis of reliability-centred maintenance as a management tool, ICOMS Conf, Wollongong, 2000 13. Al-Najjar B, Total Quality Maintenance: an approach for continuous reduction in costs of quality products, Journal of Quality in Maintenance Engineering, Vol 2, No.3, pp 2-20, 1996 14. Collins J C and Porras J I, Built to Last: Successful Habits of Visionary Companies, Century/Random House, London, 1994 15. Sherwin D J and Bossche A, The Reliability, Availability and Productiveness of Systems, Chapman and Hall, London, 1993 16. Sherwin D J. and Al-Najjar B, Practical models for condition monitoring inspection intervals, Journal of Quality in Maintenance Engineering, Vol 5, Number 3, pp 203-221, 1999 17. Christer A H, Delay time models of reliability of equipment subject to inspection monitoring, Journal of the Operational Research Society, Vol 38, pp 329-334, 1987 18. Sherwin D J. Opportunity maintenance, based on age renewal and including recursive effects, IFORS Conference, Beijing, August, 1999 ( A revised and expanded version of this paper was presented at ICOMS 2002) The author may be contacted at: djs321@lycos.com
Click here to enquire..

Applying PMBOK to Shutdowns, Turnarounds and Outages	Conference Communication
Applying PMBOK to Shutdowns, Turnarounds and Outages Bernard Ertl Vice President, InterPlan Systems Inc. Abstract Turnarounds have a distinctive character of their own and cannot be managed as if they were just another sort of ‘Engineering, Procurement and Construction’ (EPC) project; a ‘turnaround-specific’ methodology of management is needed. Drawing on the philosophy of the Project Management Institute’s Guide to the Project Management Body of Knowledge (PMBOK®) the author reviews the special problems presented by process plant turnarounds and also their particular management needs. FOREWORD Shutdowns, turnarounds and outages All the major process industries (refining, petrochemicals, power generation, pulp and paper, etc.) have their own nomenclature for maintenance projects. For the purposes of this document, ‘turnaround’ is intended to encompass all types of industrial projects for existing process plants including – • Inspection and testing • Shutdowns • Emergency outages • De-bottlenecking projects • Revamps • Catalyst regeneration, etc. i.e. all those instances when an operating plant must be shut down until the work is completed and then restarted, thus ‘turning around’ the unit or plant. In this paper ‘turnaround’ is also intended to refer to the entire span from pre-turnaround preparations, to shutdown, to execution and finally to start-up. A TURNAROUND-SPECIFIC MANAGEMENT METHODOLOGY The discipline of project management enjoys different states of maturity across different industries, the construction industry probably enjoying the greatest maturity in this field while the software development/IT industry is probably enjoying the greatest growth in maturity at this time. In the process industries the maturity of the project management discipline as regards turnarounds is still very poor, stagnant at best. There appears to be little if any development of, or dialogue about, the discipline within the field. Turnaround failures (budgets blown by millions of dollars, target dates missed by days) are still as prevalent as ever. The same mistakes are being repeated over and over. The main problem is that turnaround managers continue to treat turnarounds as if they were ‘Engineering, Procurement and Construction’ (EPC) civil projects and hence apply an EPC-centric project management methodology. One of the greatest challenges to turnaround managers is realising that turnarounds are different from EPC projects. They have their own unique characteristics and demands. They require a specialised project management methodology. This document is intended to spark a dialogue for developing a turnaround-specific management methodology. Background The Project Management Institute (PMI) has published A Guide to the Project Management Body of Knowledge (PMBOK®) which identifies and describes the subset of the project management discipline that is applicable to most projects most of the time. It provides a loose guideline for structuring a project management methodology, and describes its primary purpose as being – ‘....to identify and describe that subset of the PMBOK that is generally accepted. ‘Generally accepted’ means that the knowledge and practices described are applicable to most projects most of the time, and that there is widespread consensus about their value and usefulness. ‘Generally accepted’ does not mean that the knowledge and practices described are or should be applied uniformly on all projects; the project management team is always responsible for determining what is appropriate for any given project.’ It is up to turnaround managers to evaluate the applicability to turnarounds of their specific project management methodology. Many turnaround managers do not have a formal background in project management and have never studied the PMBOK. As a result, there has been little discussion in professional circles of the proper application of the PMBOK to a turnaround-specific management methodology. Most turnaround managers employ an EPC-centric project management approach to turnarounds. In order to analyse the applicability of this approach, we first need to understand the important differences between turnarounds and EPC projects. Then we shall have the proper basis for evaluating a turnaround-centric project management methodology in accordance with the PMBOK. Important differences between turnarounds and EPC projects There are significant differences between turnarounds and EPC projects (see Table 1), which are worth exploring. Because the scope is only partially known when execution begins, turnarounds demand much stricter scope management controls. A constantly changing scope (and schedule) means that the baseline schedule is a useless measuring stick. As it is the entire basis for measuring and tracking EPC project performance, it is clear that a different paradigm is required for turnarounds. A changing schedule and changing manpower requirements make resource levelling, a popular tool for EPC projects, counter-productive for turnarounds. This issue is explored in greater depth in the Resource Levelling Vs. Critical Mass white paper which can be found on our website (www.interplansystems.com/html-docs/resource-leveling-critical-mass.html). The compressed work basis for executing turnarounds means that all team members have less time to analyse and react to changing priorities. Problems that go unchecked can significantly affect the likelihood of reaching time and budget goals. As a consequence, there is a much greater need to use the schedule to drive the project execution (whereas it is sometimes used mostly as a contractual tool in EPC projects). It is critical for all schedule and progress information to be highly visible, timely, comprehensive and accurate. With these distinctions in mind, we can now explore the tenets of the PMBOK and start working towards a turnaround-specific project management methodology. PROJECT SCOPE MANAGEMENT One of the greatest challenges in a turnaround is scope management. This is true for virtually all phases of scope management as outlined by the PMBOK, viz. Scope Planning, Scope Definition, Scope Verification and Scope Change Control. Scope planning, definition and verification Unlike EPC projects – which usually have a well defined scope established with long lead times before the project execution phase – it is common for turnaround scopes to be changing up to the last minute before project execution. There are several factors contributing to this – • Market conditions (plant profitability) can cause variability in considerations for the budget (requiring scope adjustments), window (squeezing or relaxing the time frame available for executing the project) and start date (which may affect the decisions on what scope to include, the ability to plan the work, or material availability). • Planning input is usually derived with input from Operations, Inspection, Safety, etc., and Operations may continue to identify potential scope for the turnaround until the last minute. • The availability of specialised tools, materials, equipment and/or resources may affect decisions on how to approach portions of the scope (i.e. plans may need adjusting to accommodate a different method or scenario to accomplish the same goal). One result of this is that turnaround budgets are rarely based upon a complete, detailed plan. They are often based upon conceptual estimates, extrapolations of past turnarounds, or on incomplete planned scopes that are compensated with a large allowance for contingencies, and it is therefore necessary to review the cost estimate for the final approved scope and make sure it is covered by the approved budget or authorisation for expenditure (AFE). If not, either the scope should be culled where possible (or failure to meet the budget will be pre-determined) or the budget should be adjusted to reflect the plan (not always politically viable). Scope change control In a turnaround, the scope will change - sometimes dramatically. As equipment is opened, cleaned and inspected the extent of required repairs can be determined, planned, costed and either approved or tabled for a future window of opportunity. Every add-on to the schedule should be processed with a defined procedure for evaluation/approval.An additional challenge is presented in companies where the existing culture allows operators to direct work crews (or get supervisors to direct work crews) to perform work that for one reason or another was not included in the approved project scope. The only solution is to change the culture to respect the defined procedure for add-on approval. Operators and supervisors or superintendents must buy in to the add-on approval procedure and field hands must be directed to work only on approved scope as directed by their supervisors. Where this is not possible (or there is a ‘work in progress’) it is imperative to at least document and account for these unapproved jobs, where performed, so that progress tracking (earned value) may give a meaningful impression of the productivity of the field work. Management needs to exercise care when evaluating add-on repair scope, in order to ensure that existing resources, productivity and time can accommodate the work (where the repair scope is not operationally critical or safety critical). Management can end up in a position of balancing the impact of add-on repair work against the culling of original scope where resources become constrained on work that is non-critical (from both a time and operational or safety point of view). It is desirable to classify scheduled or progressed activities according to two main criteria, viz. (a) • Approved Scope • Unapproved Scope • Cancelled Scope and (b) • Original Scope • Add-On Scope PROJECT TIME MANAGEMENT One of the most obvious signs of the low maturity in turnaround project management is the state of the planning and scheduling that is intended to form the foundation of the management process. A successful turnaround management methodology must set a high standard for the planning and scheduling to be successful. Activity definition The planning of activities that are overly broad in scope is one of the biggest obstacles to using a schedule for any meaningful purpose. They – • are difficult to estimate with confidence, • can mask details that the planner neglected to consider, • preclude a detailed critical path analysis where more detail may allow refinements in the logic, • detract from the accuracy of progress estimation (estimating % complete is more difficult). Because of the compressed nature of turnarounds, there is a very small window available for recording and processing information on progress in order to generate updated schedules for the next shift. The greater the detail in the activity definition, the less thinking or guesswork is involved in assigning progress to the defined tasks. Activities must be clearly defined, and should be measurable. This means anyone should be able to determine whether a particular activity (as defined) is in progress, or completed. Activities must be defined every time there is a break or change in work content, or changes in the work crew. Scheduling It is of paramount importance to understand that – unlike EPC projects where a baseline schedule is often used as a firm contractual commitment – a turnaround schedule should be considered a guideline tool to drive the execution of the work. This understanding is fundamental to developing a successful turnaround management methodology. Turnaround managers have a lot of discretion with regard to scope management in schedules. While there will be portions of the scope aside from the critical path work that must be executed within the instant project, a significant portion of the scope may usually be postponed to future turnarounds or maintenance opportunities. As priorities shift, depending upon the scope of add-on repair work and resource constraints, managers need a turnaround schedule that offers flexibility in managing the non-critical work. Baseline schedules (other than for critical and near-critical paths) are meaningless for turnarounds once they start. For turnarounds, it is expected that as inspections are performed, a changing scope (and therefore priorities for constrained resources or non-time-critical work) and often poor schedule compliance (due to unavoidable circumstances) will force the schedule for non-time-critical work to change from update to update. Because of the dynamic nature of turnarounds it can be counter-productive to employ soft logic and resource levelling schemas that smooth the schedule. Both techniques are designed to produce a static plan for execution that is not practical for turnarounds. Soft logic will necessitate constant time-consuming changes and updates to maintain a meaningful schedule once deviations from the schedule occur (and this is expected in a turnaround). Resource levelling schemas that alter a hard logic schedule will introduce multiple problems within a turnaround context. (See the white paper Resource Levelling or Critical Mass? on our website: www.interplansystems.com/html-docs/resource-leveling-critical-mass.html). It is instead preferable to maintain a schedule for critical and near-critical path analysis and to allow field supervision discretion in directing their crews on non-time-critical work according to changing priorities and circumstances. At every update turnaround managers should monitor progress trends, productivity/earned value and scheduled resource requirements, to ensure that sufficient time and resources are available to complete the non-time-critical work within the span of the critical path. PROJECT COST MANAGEMENT Turnarounds are notorious for over-running the budget. Part of the problem, as mentioned earlier in the Scope Planning section, is that budgets are rarely based upon the detailed plans and estimates for the scope. Many turnarounds have failure predetermined! Should budgets be based upon a detailed plan (or at least cover the estimate for it) turnaround managers have a reasonable basis for managing costs according to the formula used in EPC projects, as laid out by the PMBOK, viz. – • influencing factors that create changes to the cost baseline to ensure that changes are agreed upon, • determining that the cost baseline has changed, • managing the actual changes when and as they occur. At the end of a turnaround, the final scope of the execution usually encompasses several categories of work, viz. • known scope (planned or estimated) • anticipated repairs (may or may not have been planned or estimated) • unanticipated repairs (not planned or estimated) • unauthorised work (not planned or estimated) • cancelled work (planned or estimated but culled during execution). So, prior to execution, the turnaround manager may have a set budget covering known scope, anticipated repairs and some contingency allowance to cater for possible remaining items. Because most indirect costs (though not necessarily material costs) are keyed off the direct labour costs, the key to successful cost control in a turnaround is execution control (keeping resources productive) and scope management (balancing add-ons against non-critical work). Earned value management In most cases, it is very difficult to obtain a meaningful earned value analysis in a turnaround. There are several problems that conspire to frustrate the system: • The process for capturing and approving actual hours usually lags behind the progress updates by at least one shift, if not two or three. • The application of the correct work order and cost code number on timesheets is poor • Unauthorised work is charged to existing work orders and cost codes but not captured for planning and estimating. Where earned value analysis is conducted, it may be most meaningful to compare numbers analysed by resource type instead of by work order or cost code. In this fashion, managers may have some measure of the productivity of the resource relative to the schedule. PROJECT QUALITY MANAGEMENT According to the PMBOK, project quality management entails several aspects, viz. – • quality planning, • quality assurance, • quality control. Of these, quality assurance and quality control are usually well defined (and in some cases, government regulated) for safety concerns in turnarounds. Quality planning, on the other hand, is an issue that is usually poorly addressed. It is recommended that organisations employ a system for capturing and improving plans and estimates for recurring jobs from turnaround to turnaround. This entails benchmarking during execution and a follow up review post-execution to update the system. Too often, the follow up review never occurs and it is left to chance that the planner will remember necessary details the next time a turnaround involving the same unit is planned. In many cases where the preparation time for planning a turnaround is compressed or inadequate there is a better than average chance that turnaround plans based upon templates (or historical plans) will suffer from cut and paste syndrome and not receive due consideration for customising to the instant situation. The best solution for mitigating these problems is to use a planning system like eTaskMaker®. PROJECT HUMAN RESOURCE MANAGEMENT Most turnarounds involve managing a large contract work force to execute the project. The dynamics of organisational planning and staffing are usually well understood. The industry is mature enough for project roles and responsibilities to be well defined. There is a mature support industry of specialised and general contractors to supply the necessary human resources. Probably the greatest challenge that turnarounds present in comparison with EPC projects is the management of the human resource. This is true at a macro (overall staffing levels) and micro (delegation of work to labour pools) level. Most turnarounds staffing levels can be represented with a bell curve. Staffing levels for specialised skills are increased slowly as units are blinded and vessels opened. Staffing levels are reduced as repairs are completed near the end of the turnaround. Managers should analyse staffing levels versus schedule requirements frequently (in some cases daily) to ensure that sufficient manpower is available to complete the bulk of non-time-critical work within the span of the critical path while also demobilising excess manpower to control costs. It is critical to the successful execution of a turnaround for field supervisors and superintendents to foster a co-operative teamwork spirit with regards to managing the existing labour pool. Where supervisors and superintendents are held accountable for meeting individual schedule and progress goals, competition for (and hoarding of) skilled labour can occur and jeopardise the project success. Supervisors and superintendents should bear equal responsibility for meeting the overall schedule and progress goals. PROJECT COMMUNICATIONS MANAGEMENT One of the single most important aspects of successful turnaround management is communication. Because of the compressed time frame, there is less time available to everyone in the turnaround team to overcome the problems caused by poor communication. Communications planning Many turnaround organisations do not have a communications plan outlining team members’ information needs, delivery schedule and distribution system. Ad hoc reporting does not provide the necessary foundation for maintaining high visibility of the project to all stakeholders. A proper communications plan should include – • Executive Management (summary schedule, progress) liaising with Turnaround Management (scope, schedule, progress, manpower) • Planning/Scheduling (scope, schedule, progress) liaising with Inspection (schedule, progress) • Operations (scope, schedule, progress) liaising with Safety (scope, schedule [permit requirements]) • Warehouse (scope, schedule) Information distribution Since turnarounds are so dynamic, information needs to be updated every shift if visibility and control are to be maintained. In order to help field supervision stay on top of changing schedule priorities, it is recommended that complete schedule updates be initiated just before the end of every shift so that they may be disseminated to the field at the start of the next shift. Without this, the schedule will quickly become meaningless as a tool for managing and driving the project scope and execution. Stakeholders in turnarounds are always pressed for time. It is recommended that all disseminated information conform to standard report formats, familiarity with which will enable team members to read and digest the information quickly and will minimise the potential for misinterpretations. Performance reporting A turnaround project should not be analysed in the same manner as an EPC project. Their dynamics and characteristics are different. Baseline schedules that drive EPC project analysis are relatively meaningless for turnarounds after the first couple of shifts. Turnarounds require specialised metrics for analysis, e.g.– • Summary progress attainment curves for measuring, tracking and trending against earned progress. • Critical mass (see our website – www.interplansystems.com/turnaround-project-management-primer/critical-mass.html). PROJECT RISK MANAGEMENT Turnarounds usually entail a high degree of risk. Because the scope of work is only partially known, managers must prepare for the possibility that the effort to clean or repair equipment may exceed estimates and expectations when the equipment is opened and inspected for the first time. In general, it is not practical to attempt to model all potential risks within the project schedule. There are virtually infinite possibilities for required repairs on the more complicated pieces of plant equipment like compressors, heaters/furnaces, towers, etc. It is recommended that risk analysis be considered for critical and near-critical path work in the schedule because such work carries the greatest likelihood of affecting timely completion. Managers must balance the cost of maintaining any specialised parts or materials for identified risks against the cost of a possible schedule delay should such parts or materials need to be procured at the last minute. CONCLUSION Turnarounds are very challenging and dynamic high performance projects. They have many unique characteristics that differentiate them from other types of projects. The EPC-centric approach to project management does not work very well for managing turnarounds. It is hoped that this paper may prompt the reader to consider employing a turnaround-specific methodology for managing turnarounds. Considering the stakes involved, it is high time that industry adopted a more mature approach to this task. Quotes: *Operators and supervisors or superintendents must buy in to the add-on approval procedure and field hands must be directed to work only on approved scope as directed by their supervisors.* It is of paramount importance to understand that a turnaround schedule should be considered a guideline tool to drive the execution of the work. This understanding is fundamental to developing a successful turnaround management methodology. *It is recommended that risk analysis be considered for critical and near-critical path work in the schedule because such work carries the greatest likelihood of affecting timely completion.*
Click here to enquire..

Improved CMMS and Asset Management Systems - But do they lead to success?	Conference Communication
Improved CMMS and Asset Management Systems – But do they lead to success? Len Bradshaw Editor,The Maintenance Journal, Victoria, Australia PART 1 IMPROVED CMMS AND ASSET MANAGEMENT SYSTEMS Computerised Maintenance Management Systems (CMMS) evolved in 70’s and 80’s as a means of managing maintenance activities. In particular they were created to help manage all aspects of managing, planning, controlling, requesting, recording, reporting, and analysing maintenance activities. The basic capabilities of a CMMS The basic requirements of CMMSs are shown in Figure 1 and listed below, viz. – • Providing a data base for the assets: Inventory of the assets; Bills of materials; Asset register or technical database; Asset safety, isolation, permits and regulations; History records. • Storing descriptions of maintenance activities: Corrective maintenance procedures and repair details; Planned maintenance procedures, frequencies, etc. • A means of maintenance requesting and responding:Requesting maintenance assistance; Prioritising maintenance requests; Initial processing of maintenance requests; Checking resources required; Backlog files; Responding appropriately to emergency/urgent requests; Providing feedback to requesters. • Preventive maintenance: Triggering preventive maintenance tasks; Identifying and triggering opportunity maintenance. • Maintenance work scheduling, issue and control: Resources required – people, parts, tools, etc.; Scheduling work; Resource balancing; Work lists; Work orders – development and issue, permits, access, isolation; Controlling work in progress; Response to further work required. • Recording and analysis of work performed: Feedback and monitoring of work performed; Closing selected data to history files; Analysing history data;History, performance and cost reports; Closing the loop. Improved CMMSs There have been dramatic improvements in the ease of use, speed, and functionality of CMMSs. In fact, the improvements go far beyond this. CMMSs now provide much more than a basic maintenance management system. They have improved capabilities such as – K Integration and interfacing • Direct linkage to stores, purchasing, costing and production management systems. (Such interlinked systems go beyond being simply CMMSs and are then Enterprise Asset Management Systems [ EAMSs]). • Linkage to other specialist systems such as those concerned with project management, energy management, and analytical software, etc. • Ability to access and disseminate diagrams and pictures, CAD/CAM and videos, using document imaging, etc. Direct linkage to condition monitoring systems – providing, if required, condition data from a plant item to anywhere in the world. • Direct linkage to control systems, production control data, etc. K Communication, data collection, data transfer • Portable data collection and data transfer systems, etc. • Bar coding, stick-on data buttons, transponders, electronic tagging, etc. • Radio paging, data transmission, telemetry systems. • Internet, intranet, web-based capabilities. • Links to global positioning systems (GPS). • Incorporation of geographic information systems (GIS). Integration and interfacing of CMMSs A CMMS is a mechanism for communication not just within the maintenance department but also with other departments and possibly even other organisations (see Figure 2). Chris Cooper¹ has suggested that – K ‘A true EAMS is one that provides maintenance functionality but is also a fully integrated module of an enterprise system. The integration will then be part of the overall software design and immediate updating of files takes place rather than data being passed between system modules’. K ‘A third-party EAMS will usually require data to be passed between systems and is therefore interfaced not integrated’. The benefits achievable by a CMMS are very much dependent on the extent of the integration of the maintenance management system with other sections of the organisation. The selected system must not only satisfy the planning, control and information needs of the maintenance department, but must also provide the data/information flow to and from the other sections of the organisation. The following are some informative quotations regarding the extended use of CMMSs and EAMSs – K ‘While our ability to control specific transactions and work flows through information technology continues to increase, how effective are we at making tactical decisions based on the mounds of data we collect? Information may be an asset, but in copious quantities it can literally choke the decision-making process. We might be impressed by the number of reports and inquiries within our enterprise applications, but can we effectively use these tools? Despite all their underlying data, do our applications present the information we really need in a manner that we can effectively use in our decision-making processes?’ Tom Singer, Principal, Tompkins Associates² K ‘What if the loop could be closed – what if your factory-floor could ‘talk back’ to your CMMS in real time? This is the essence of predictive maintenance or condition-based monitoring, and the future of asset management and CMMS. Using the data inherent in almost all plant-floor automation and control systems, coupled with advanced Condition-Based Monitoring (CBM) technologies you can transform the CMMS into a truly responsive predictive maintenance system (with) – • Computerisation of maintenance scheduling, spares procurement, plant equipment databases and so on, using leading CMMS technologies. • Automated data collection: reducing laborious and costly routine data collation by drawing-in data straight from its factory-floor source. • CBM: applying advanced CBM schemes to protect investment in major plant items.’ Rockwell Automation, ‘Towards predictive maintenance – Listen to the factory floor’ K ‘Internet companies are investing millions of pounds and dollars building virtual market places called portals. There are industry specific portals and general portals but they all operate in much the same way. If you wish to send out a tender to a number of suppliers then all you need do is send the tender to a portal you subscribe to and they will circulate it to appropriate suppliers. The suppliers reply to the portal operator who forwards the responses to you for consideration. When you decide on the best supply contract, you place your order through the portal operator and this in theory cuts down the time and effort you spend in tracking down the most competitive deal.’ Philip Taylor, Commercial Director, Engica Technology Systems³. Communication, data collection, and data transfer Barcoding – of parts, work instructions, personnel, equipment and tools. This allows less paperwork and data entry workload, provides more accurate reporting, and can be used to verify the time and date of an activity. Stick-on memory buttons – small stainless-steel-encased electronic buttons provide an alternative to bar-coding that can be used in wet and dirty environments (they only need electrical contact with the hand- held data collector). Buttons are also available that measure temperature and have the ability to store maintenance data (i.e. maintenance instructions can be transferred to the data logger when contact is made with the button). Transponders – perform a similar function to the above but do not require direct electrical contact. The data-logger sends a radio signal, that energises the transponder, and which allows transfer of data to or from the transponder. Hand-held data loggers and Palm devices – are small computers that are transportable, often pocket sized, and can be used in most maintenance environments. K They can be programmed for PM routines or inspections, providing the tradesman or technician with details of the asset and the maintenance work; K The inspection person can respond to each prompted task by using a bar-code reader or keying, into the data-logger, numbered codes from fault or response lists; K Some can be fitted with measurement probes for temperature, pressure, vibration levels or electrical measurements; K At the end of a day of inspection activities the information can be transmitted to the main CMMS by direct line, radio or telephone. K They may also be used in conjunction with a GPS which, for widely distributed assets (roads, power poles, etc.) or for mobile assets (long distance haulage vehicles, etc.), may be a very useful facility, enabling accurate location of the point at which maintenance is being performed or the point at which a fault is being reported. There are also Palm devices that link to Geographic Information Systems (GISs) and provide GIS displays, maps, etc. K Plant operators may use machine or wall-mounted data loggers to considerably improve the quality of data, and the speed of data collection (e.g. for downtime data collection). K Those with bar-code reading facilities are used in a stores environment, and used by the tradesman as a means of recording his arrival at a particular asset and linking that asset to a maintenance activity. PART 2 WE NOW HAVE GREAT SYSTEMS BUT WHERE ARE THE GREAT RESULTS? Even with modern CMMSs and EAMSs we still find that there are implementation failures, or perhaps that the impressive range of functions is under utilised (the expensive electronic filing cabinet). These improved systems, while perhaps helping to create satisfactory performance in managing maintenance, do not necessarily lead to your maintenance organisation becoming among the ‘best’ of its kind. What leads to dissatisfaction, satisfaction or success in managing maintenance? Motivational theory Let us first examine one of the better known theories on motivation, Herzberg’s Motivation and Hygiene Theory⁴. Herzberg considered that the factors shown in Figure 3(a) – Achievement, Recognition, Work Itself, Responsibility, Advancement and Growth – are the true motivators. The factors shown in Figure 3(b) are the ‘hygiene’ factors which if not up to a certain level or standard cause significant dissatisfaction. However, once those basic needs or standards are reached further improvements to those hygiene factors will not, on their own, create significant or sustainable levels of motivation in the workforce. So if we create an interesting and varied work environment, where we are proud of what our work group is able to achieve, then this is a situation in which the workforce is likely to be motivated. Among the hygiene factors ‘Salary’ often generates questions of motivation. If our salary is already at a reasonable level (and therefore not a source of major dissatisfaction) and we receive a big increase in salary it may generate motivation for a short period but we quickly accept the new salary level as the expected norm and it then is no longer a motivator. Applying motivational theory to the use of CMMSs and EAMs Selection and implementation of maintenance management systems The failure to properly select and implement a CMMS is often caused by paying insufficient attention, during this process, to human factors. The way we select and implement a CMMS or EAMS can be a major source of dissatisfaction (see Figure 4). Consider the following extract from a recent paper by Labib⁵. ‘It appears that there is a new breed of CMMSs that are complicated and lack basic aspects of user-friendliness. Although they emphasise integration and logistics capabilities, they tend to ignore the fundamental reason for implementing CMMSs, i.e. reducing breakdowns. These systems are difficult to handle by either Production or Maintenance. They are more accounting- or IT-oriented rather than engineering-based’. K Are such CMMSs chosen to meet maintenance objectives and to meet the needs of maintenance personnel? No. K Were maintenance personnel actively involved in the selection process? It’s not very likely. K Is the training on these difficult-to-use systems going to be easily achieved? No. K Are your maintenance personnel going to be motivated to contribute enthusiastically to the implementation process of this imposed system? It’s not very likely. K Is this system going to be a source of dissatisfaction? Very probably. Such systems will tend to be a major source of dissatisfaction within your maintenance workforce. In one organisation’s successful implementation of a CMMS the reasons given for their success were as follows – • Forward planning, which meant that the project was not going to be forced on to the personnel involved. • The personnel were asked for opinions and ideas, and at all times were involved in the introduction. • Section heads were kept well informed of progress. • Training was conducted in a manner which helped people accept the change. • Training was conducted on ‘home ground’ and people were more involved. • Management had continually shown its support and desire to meet, where possible, the individual needs of everyone who used the system, e.g. design for individual problems. • Problems were diagnosed and corrected as soon as possible. It is clear that successful implementation of a CMMS is heavily dependent on the following human factors issues – • Adequate consideration of human factors. • Involvement of persons affected by the new system in the design, specification and implementation process. • The provision of adequate training for all levels of personnel in the system objectives and system operation. Using the CMMS or EAMS Functionality and support for the CMMS: As with the selection and implementation process the poor functionality of the CMMS or EAMS and poor support for it can be a major source of dissatisfaction. However, as long as the functionality and support are reasonable the dissatisfaction will generally be overcome. Very good functionality and support can start to contribute to motivation but does not play a major part in this. It will not necessarily lead you to excellence in maintenance. Consider one of the most recognised sites in the world for excellence in maintenance planning and maintenance management. Their CMMS was adequate for more than ten years but far from the best in terms of functionality and ease of use. You can be the best even if you do not have the best CMMS or EAMS. Their path to being the best lay in using the remaining factors shown in Figure 5 plus those shown in Figure 6. Clearly defined and policed rules of use:Involving as many people as possible in defining the structures, systems, use and responsibilities and then sticking with the chosen methodologies. For example – K ‘All maintenance requests must be documented before job start. Even urgent work will require the creation of a ‘quick work order’ prior to work on such jobs’. K ‘This production area will have a maintenance co-ordination meeting each Wednesday, in this room, involving these people. The group will discuss planned preventive and corrective work for the following week. The production department cannot, at any point beyond that meeting, refuse access for agreed planned work, except in the case of clear emergencies, and only then if such refusal of plant access is in writing from production management’. In short, set the rules and make sure those rules are followed. Monitor and review: Ensure there is a system in place to monitor CMMS usage and the results or achievements made via the CMMS. Again, set a regular review process and review period. Set performance parameters for the CMMS relative to its use and outcomes. For example – K Ratios of work issued to work completed. K Access rates to the various CMMS modules by maintenance personnel. K Quality standards for maintenance history data. The maintenance planner: I may be old fashioned but I still believe that in medium-to-large maintenance groups the key to success is a full-time dedicated and motivated planner. A poor quality planner will be a major source of dissatisfaction whereas a good one should be a salesman for the CMMS or EAMS and a key motivator for others to work with the system rather than against it. If you speak with the planner and he or she is whinging and complaining about the CMMS you have no hope of higher level success. People issues The final and most important set of factors is shown in Figure 6. These ‘people issues’ may contribute to great success in using CMMS or EAMSs or, conversely, may create great dissatisfaction. The first two factors, viz. Recruitment and People Resource Levels, require little comment. If your organisation has recruited well then your group of motivated team players will make your system work well (even if it is a pig of a system). If, however, your organisation has recruited badly (with a poor work culture and ‘don’t give a damn’ mentality) then even if you have the best CMMS or EAMS in the world it will never be successful. Similarly, if you have insufficient maintenance personnel the people are pulled off PMs to attend to failures, which lead to more failures and fewer PMs – the downward spiral to fire fighting. The CMMS or EAMS will help make better use of your limited people resources. It will also help to identify the extent of the problem (incomplete PMs, backlog levels, etc.). Teams: I am a fan of teams whether they are maintenance teams dedicated to a particular area, or mixed maintenance-production teams. Teams that are created in the right way and made up of motivated team players are great. They can bring together all of those true motivating factors of Herzberg’s (see Figure 4[a]). In high level teams it is very much the team members who not only collect history for the CMMS but will also be using the history for improvement strategies. They recognise the value of the CMMS as a management and decision making tool that they interact with every day. A US company, Advanced Software Design, stated the following – ‘For a product to be fully and willingly utilised, it must offer value to the person who must enter data; satisfying management’s information needs is not adequate reason to ensure diligent usage of the product. Value to the technician/craftsman invariably falls into one of several areas, viz. • It makes their job easier. • It allows them to do their job better • It reduces tedious tasks. • It makes their job more interesting. • It increases their value and therefore their probability of future higher earnings’. A good maintenance team utilising a good CMMS or EAMS can get close to achieving the above conditions. Trust: How much time and money is wasted because of the lack of trust? K Are CMMSs, and particularly EAMSs, in your organisation to provide effective maintenance and asset management tools? Or are they there to check on each employee as to the value of his work and the dollars he costs? K As an employee do you trust your managers and accountants to use the information you input to the system in a fair and reasonable manner? K As an employee how often do you fear reporting the truth to your CMMS? Without trust, when in an atmosphere of blame and penalties, we use the CMMS or EAMS to play games. K ‘Underworked – you must be joking. Just look at my daily job sheets – I have been busy every hour of every day for the last twenty years’ (Must be great planning !). K ‘I am pleased to announce that this team has attained its new performance targets’ (But nothing has actually changed! ‘Hey, what percentage figure do you want? We will supply the data that will produce that figure for you!’) K ‘Last quarter this team recorded the lowest backlog figures for this company. Unfortunately management used this as evidence our team is over-resourced and has moved two of our guys to another team.’ (I wonder just how high those backlog figures will be in the future!) Leadership: If your organisation is to be the best at maintaining, managing and using your company’s assets it requires good leadership. In terms of a CMMS or EAMS this means you need someone at the top of your organisation who will champion the selection, implementation and on-going use of the CMMS or EAMS. Hugh Blackwood of Alcoa’s Mt Holly plant has stated⁶ that good leaders – • Create a sense of urgency - this is not ‘programme of the month’. • Understand the plan so that it can be shared with others. • Communicate with the people they work with (i.e. ‘walk the talk’). • Encourage people into broad-based action. • Focus - they begin generating short-term results. • Lead - success depends on it! It is clear that successful implementation of a CMMS or an EAMS is heavily dependent on people issues. ‘Working in a continuous flow manufacturing environment, I’ve witnessed tens of millions of dollars in capital investment at our facility over the past four years. We are now dutifully equipped with all the latest bells and whistles, from automation to expanded PLC control and process monitoring. However, we still cannot track downtime causes, perform root cause problem solving, locate spare parts, or follow standard work practices. Because of this, we continue to flounder. ºWe’ve been given a Corvette but have yet to get our driver’s license! Editorial letter, Maintenance Technology, October 2000. REFERENCES 1. Cooper C, Holistic RCM, setting a new corporate strategy for maintenance management, Maintec 2002, UK 2. Singer T, Information engineering – the search for business intelligence, Plant Engineering, November 2001 3. Taylor P, Impact of computer technology on maintenance, Maintec 2001. UK 4. Herzberg F, One more time: how do we motivate employees? Harvard Business Review, January 1968 5. Labib A, CMMS, black hole or black box, Maintenance Journal, February 2004 6. Blackwood H, Five years of changes at Alcoa’s Mt Holly Plant - what have we learned? International Maintenance Management Conference, Australia 2002
Click here to enquire..

Automated Trouble Shooting	Conference Communication
Automated trouble-shooting Alan Finn Automated Reasoning Inc. Abstract Automated trouble shooting of problems with plant or products, using diagnostic software that may be on site or may be accessed remotely via the internet, can be highly cost-effective. It is explained that while many diagnostic engines are case-based, relying on learning from prior case history, which carries inherent limitations, a more sophisticated system – which also employs model-based reasoning (as developed and marketed by the author’s company) and in which the integrated method of diagnostic decision-making closely resembles the human thinking process – can be much more efficient. The logic of such a system, its capabilities, the benefits it can bring and the extent of its current industrial application are reviewed. INTRODUCTION Automated trouble-shooting by diagnostic software had its roots, some fifteen to twenty years ago, in medical fields and, later, in diagnostics for complex military equipment. Although initially designed to carry out medical diagnostics on the human body, difficulties were met with US product liability legislation. A later approach by the Pentagon for deployment of diagnostic software on the equipment in nuclear submarines, to facilitate troubleshooting in the minimum time while on patrol and to minimise the inventory of on-board spare parts, proved very successful. Due to the increasing sophistication, power and rapidly falling costs of such software, and its ability to be used remotely over the internet, it can now provide the basis of highly cost-effective solutions to the problems faced by diagnostic engineers. Expert system software, using diagnostic models and reasoning engines, is reaching its full potential for remote diagnostics through the world-wide web, providing menu-driven, fault-solution self-help to remote locations. There are huge cost savings to be made using web-centred equipment self-diagnostics and predictive maintenance and by enabling local self-help over the web, rather than sending out field service engineers to carry out on-site fault-finding. A SYSTEM WITH MODEL-BASED REASONING Diagnostic engines come in several types but are commonly case-based, relying on learning from prior case history, with all its inherent limitations. A much more efficient and sophisticated system underpins, for example, my own organisation’s products, ‘Intelligent Computer-Aided Logic (ICAL)’ and ‘Intelligent Computer-Aided Trouble-shooter (ICAT)’, which employ model-based reasoning (MBR), with case-based reasoning (CBR) added. The diagnostic and prognostic process we use (finding critical, developing, faults before they manifest themselves as equipment failures) employs ICAL to build a model of the equipment which is to be protected from long fault-induced downtimes and then the ICAT run-time component is used to diagnose the problem – either remotely or locally. This diagnostic process is based on a unique combination of MBR, CBR, rule-based reasoning and probabilistic networks that operate on principles similar to neural network algorithms. This network keeps refining itself by built-in, adaptive learning, algorithms. Automated troubleshooting, or test sequencing, is based not solely on a ‘likely improvement in understanding’, but also adds-in the dimensions of time, priorities and safety issues. MBR has the capability to incorporate data – for assistance in the decision-making process – such as failure rates, the distribution of potential fault ‘blame’ on parts of the equipment under diagnosis, costs of tests, costs of parts and test run-times. This higher level of intelligence is unavailable in systems which commonly rely solely on CBR or rule-based technology. This integrated method of diagnostic decision-making closely resembles the human thinking process and utilises historical cases in two ways, viz. 1 as input to the adaptive-learning algorithm to refine the MBR, 2 for reasoning by analogy, along the principles of ‘classic’ CBR (‘select the nearest-neighbour case’). CBR alone cannot yield ‘good’ diagnostic results in the early stages because there are no cases for its foundation, unlike its MBR counterpart. But it can refine the diagnostic model with actual case-data and accumulated failure histories. A classic neural network is relatively weak in large-scale diagnostics because the complex inter-relationships that need to be learned require a very large and representative statistical sample of historical cases. Combining the advantages of each technology produces optimal diagnostics for fast deployment and accuracy of diagnostic results. Furthermore, ICAL’s knowledge base is not limited to a single fault assumption, as is the case with most other ‘expert’ systems, and ICAT, the run-time diagnostic tool, can solve even multiple fault scenarios, tracking them down, one by one. Some systems claim a type of self-learning capability, learning and improving the efficiency of the underlying diagnostic engine from the results of each diagnostic case. ICAL’s ability for self-learning modifies probabilities in the underlying algorithms and also improves test strategies, reducing the time to fault isolation and learning automatically each time from the results of every diagnostic case. The utilisation of the applications’ accumulated experiences, combined with ongoing automatic learning, further improves the probability of ICAT’s indicating the ‘true’ faulty component in a given ‘ambiguity group’ of suspected components – eliminating the need to maintain the knowledge base manually. As new information is acquired, ICAL with ICAT has the capability to learn about and identify suspect components, via recognition of failure trends. ICAL is capable of remembering which parts were repaired or replaced in each troubleshooting scenario. This is true even if the diagnostics session is stopped – while waiting for spare parts, for example, or while other scenarios are being handled – and then resumed. This data is used for learning and generating many types of reports which can be of great value to company management. Our own products follow the ‘test-isolate-repair-verify’ cycle. The first stage is to test towards fault isolation; the second stage is component replacement or repair instructions; the last stage repair and verification, testing for a fully functional system. Documentation can guide users, step by step, through the diagnostic process, with visual and text aids and even sounds. It presents this information through ICAT to the diagnostics user, instructing them on what to do, how to do it and when to perform various aspects of the diagnostic process. Associative hyperlinks are embedded within the text or graphics, accessing schematics, photographs, video or audio clips and even existing documentation on CD-ROM. The ODBC-compliant SQL* database facilitates linking to other databases utilising SQL and ODBC. Types of equipment, serial numbers, locations and technicians’ names are standard information which ICAT collects at the beginning of a diagnostic session. At the session end, the list of faulty components, all pass and fail test results, times and dates, and status of cases are stored in the diagnostics database, together with a free comment field in which any additional information may be added. A report mechanism allows this information to be viewed, queried, and printed as a feedback mechanism for management. Advanced filtering techniques can be utilised to produce a format of choice, which can include various statistics on failure occurrences and percentages. The knowledge base is flexible and allows those carrying out diagnoses to follow easily the test sequencing recommendations, or to choose independent actions. The knowledge base tracks and analyses all actions and is ready to support the users’ decisions. The users can navigate from one page, such as a schematic or board layout, to another page, such as removal instructions, or parts bin locations. The ICAL model developer can build its knowledge base from the following data – Where design data exists in electronic format, a common condition today where CAD systems are used, then this can be imported straight into ICAL through a simple interface, building the diagnostic model very quickly and robustly from this information source. However, the diagnostic model can easily be built manually using the modelling component. APPLICATIONS AND BENEFITS It should always be borne in mind, of course, that as indicated by the Pareto Principle, about 80% of faults are commonly occurring, simple and solved quickly. So remotely informed self-help often works very well. Our own products operate in several different modes, viz. – 1 Automatic and predictive fault-finding, using sensors embedded in the equipment under test, measuring key parameters and connected over the web to the database. This tracks faults developing over time, diagnosing them and signalling alarms; 2 Local equipment users may wish to operate “self-help” via simple menus automatically selected by the diagnostic software. 3 If (2) fails the results are passed automatically over the web to the next links) in the ‘diagnostic chain’, which are often help desks or call centres, singly or in cascade. These carry on the remote diagnostic process via more complex menus presented by the diagnostic software; 4 Finally, if remote diagnostics do not solve the problem, then the field service engineer must be called out. However, he has the results of the previous ‘upstream’ diagnoses and now has a fairly clear idea of the problem, taking with him only the spare parts likely to be required. He will use ICAT on-site, finally solving – and fixing – the problem very quickly. Model-based reasoning systems can home-in rapidly even on problems they have never seen before. By its use of the complex underlying algorithms, developed originally for the medical profession in the USA, ICAL uses prior knowledge in the diagnostic model and extrapolates that to determine the most likely faults. No longer must field service engineers carry out fault-finding on broken-down equipment only by going on-site! They are now sent out to repairs only as a last resort. Sensors, embedded in domestic equipment such as set-top boxes for digital TVs, or in high-value capital equipment such as public escalators and elevators, can track key parameters and detect pending problems even before they manifest themselves as equipment failures. Remote diagnostics are used successfully on plant and products in industries as diverse as automotive and railway passenger vehicles; in aerospace and military markets; printed wiring board assembly; recording studios; utilities; industrial process equipment; household appliances and IT equipment. The diagnostic process will address faults in electrical and electronic systems, mechanical and hydraulic components and even faults in software modules. Such remote diagnostic software can be used either in local, stand-alone mode, often preferred in military environments, but more usually remotely, over the internet, by means of which problems are continuously reported to the diagnostic system over the web in remote, ‘Predictive Maintenance’ or prognostic mode. Or user self-help, linked to the remote diagnostic engine, may be carried out, but to a fairly simple test menu. If the fault is not diagnosed by the user the problem is passed up to the next link, to the help desk or call centre. All results from the prior diagnoses are presented, increasing the opportunity for fast diagnostic success and highlighting the parts needed for the repair, the objective being ‘call-out avoidance’, removing the need for the field service engineer to be sent out (an expensive option). This creates very considerable economies for companies using service engineers, allowing them to do much more with their field service teams. The savings on constant re-training on new equipment drive most of the savings from automated trouble-shooting, together with the retention and management of all product knowledge within the organisation, even when engineers leave or retire and permit their replacements to be of a lower grade, the knowledge and experience being retained in the diagnostic software’s database. The author may be contacted via – Tel: 01420 80642 Mobile: 07967 586 151 E-mail: alan.finn@automated-reasoning.com
Click here to enquire..

Three into One does go	Conference Communication
Three into One does go Andy Davies Facilities Manager, Jet UK Ltd Abstract To speed up its response to sales demands a company making office equipment concentrated all its operations on a single site, having previously had its activities distributed over three locations. It also decided that, in parallel with this move, it would replaced its outdated and inadequate paperwork systems for controlling maintenance with a Computerised Maintenance Management System. The author describes how he brought this change about, stressing the value of careful preparation for the transition and the benefits that are now being achieved. OUR COMPANY Jet UK Ltd is an independent family run business. We manufacture quality office filing products from purpose-built 135,000 sq. ft. premises in Surrey Quays, London. The business began in 1947 and over the years had developed in size to three sites that incorporated production and warehouse storage to meet the demands of the customers. In the latter part of 2000 the company moved all three outlets into one building, establishing control over manufacturing and central storage in order to minimise unnecessary movement between sites and to maximise quicker delivery in response to the urgent demands from the sales team. THE OLD WAYS AND CHANGE The move from the three sites bought with it different concepts of working, from Production through to Maintenance, and also the need to establish a workable solution to the problem of keeping the production lines fully operational and holding the correct level of stock within the warehouse to provide a seamless route of delivery. As with most maintenance departments the new and suddenly changed working practice took quite a while to become the norm, and when to this was added the activity involved in moving the machines from the previous three sites to the new building there were certainly many different tasks that required input from the maintenance team – a task which was over and above the normal work pattern they were used to. When I joined the team in 2002 the maintenance department had started to try and establish certain criteria to improve performance within the department. The site operates for two shifts each day, 06.00 –13.30 and 13.30 – 21.00, the maintenance team providing support over the same shift pattern, the team for each shift comprising two mechanical engineers and one electrical engineer. Fault reporting and work request reporting was carried out via a paperwork system and no history archive was created for future reference. Stock and spares were purchased on a ‘need-to-have’ basis or on a large quantity ‘just-in-case’ basis and there was no stock register in place. There was also no tracking system for parts purchased and no cost monitoring on department spend. With my background in facilities and maintenance, using applications for the retrieval of information, I started to tailor the work-reporting system already in place on to spreadsheets for analysis of department performance. The work reporting system consisted of a printed work order, via which the fault and/or the work required was reported and which was hand-written and given into the maintenance office. A maintenance engineer would be allocated to carry out the necessary repairs and, on completion of the work, the engineer would then fill in an internal work order to report on action taken to rectify the fault or on the work that was required. Each week I would categorise the completed work orders and enter these on to a spreadsheet to show how many had been raised and the hours spent on each category. The categories encompassed Breakdown, Maintenance Work, Health and Safety, Project Work, Planned Preventive Maintenance (PPM) and Facilities. This gave me an indication of where the maintenance time was spent and what we needed to do to improve machine reliability. Within the maintenance workshop I introduced workplace inspections and applied this to the areas which came under our control – viz. the plant rooms and switch rooms – so that these could be maintained in a clean and tidy condition in keeping with inspection audits. No planned maintenance had been formally implemented and neither were there any task schedules written to work from. I therefore created basic PPM schedules which captured statutory inspections and recorded these onto an Excel spreadsheet. On the facilities side we recorded weekly meter readings to monitor gas, electricity and water consumption and used the figures thus derived to confirm utilities service bills. I carried out all the purchasing for Maintenance – of everything from the spare parts for the machines to the materials needed for the maintenance of the building – and logged every purchase order on to a spreadsheet to give me an indication on department spend; this would also create a target to reduce costs on unnecessary purchases. In order to accrue costs for the annual contracts I set up a database and asked all the contractors to send in the yearly costs, which gave me an overall spend for the year. Along with the management team we had been looking at a Computerised Maintenance Management System (CMMS) that would handle all of our requirements from work reporting through to stock control – a system that would be easy to use for data entry. BEST-FIT SYSTEM Over the years I had worked with many CMMSs and I looked for a system that combined all the nice bits that were really useful into one package that would suit our requirements and be cost-effective at the early stages of implementation. We chose a system that gave us asset management, stores and stock control and a PPM module which would bring the maintenance department into a new method of operation. A lot of things are said when new ideas are introduced to people with limited computer knowledge, e.g. ‘That will never work’ or ‘Never had to do it that way before’, and that is why it was so important to get a system that would be more then user friendly. The system also had to be flexible. It had to be capable of interfacing with monitoring systems if, in the future, we were to deploy these, and of linking to the production schedule in order to indicate when machines are not likely to be used, so that PPMs or reactive maintenance could be carried out. ACTION PLAN AND IMPLEMENTATION Before a new system can be introduced you need to establish a plan of how it is going to work and to prepare data to load into the new system so that it does not become a continuous task. I compiled asset lists and created a stock register on Excel spreadsheets, because these can be pre-loaded before the system is installed and will facilitate getting the system live at an earlier stage, The advantage of having all the information required to build the system is that it allows you to be able to run from Day 1 and to add data as it grows. I found this the best way of leaning how to operate the system and only needed half a day’s training from the software company. We installed a central computer that allowed the shop floor personnel to report faults or other types of work directly, and on closure of the work order it was then printed into the maintenance department. The task was then allocated to an engineer, as one became available, to carry out the necessary repairs. The work order screen comprised two pages where personnel could enter asset number, job description, trade, and priority, and the ease of creating the work order surprised even the most ardent non-users of computers. To the extent of their saying ‘It’s quicker then writing’ this was a big plus in winning them over to the new technology. The other advantage in getting the shop floor to create work orders was that I did not need to employ an administrator to work the system. When the work was completed the work order was closed down on a daily basis to ensure that the work in progress was kept to a minimum. With the help of the software company in the early days I prepared a lot of the ground work before we had the system installed, creating location lists and location levels that could be used even at different sites. A typical entry for one asset, to capture all the information, required Location Levels, Manufacturer, Suppliers, Original Cost (to capture depreciation), Year of Manufacture, Serial Numbers, Model or Type of Machine or plant (these last two items were usually the information required by the supplier to identify spares and new parts). So anyone who has not installed a CMMS can appreciate there is quite a bit of work to be done before your first report can be generated to prove investment well spent!. THE SYSTEM GOES LIVE I spent three to four months building the system, entering all the data required for a solid working system, and during that time I showed the supervisors, managers and workshop personnel what the system could do and what they could gain from it. I also got their input for improvements before it went into use. I ran the system alongside the old paperwork reporting operation so that I could fine-tune the asset entry and work reporting module, I then slowly introduced work orders produced by me into the maintenance department and trained the shop floor on how to generate their own job tickets. It wasn’t too long before the old operation stopped and every work order was being produced from the CMMS. This also produced a cost saving as the paperwork reporting and internal reports were produced on site, which meant having to store a very large quantity of old records. From shortly after we installed the system every statutory inspection was generated by it, which provided a better management of the recording of tasks done on work such as fire alarm testing, the recording of each zone tested ensuring that each was tested with the required frequency. The maintenance department is currently working on the stores stock module, entering all details and attaching known parts to the asset inventory. The reports, when run on certain assets, will be able to highlight labour costs and cost of parts used, from stock or purchased. The added benefit here is that you can easily spot high spend on one particular part and deploy a design change or improvement to eliminate or reduce the incidence of the recurrent failure. The system is now being used full time for fault logging and the personnel have recognised that it is also better, when they require new parts, to attach the request to the asset via the work order; this is costed to the asset when the part comes in and it helps me in remembering to order the part. FEEDBACK AND CONTINUOUS IMPROVEMENT It is still early days for praise or critical feedback but so far it is all good feelings from the shop floor as regards the system as a tool for reporting machine breakdowns or other tasks that require an input from the maintenance team. ‘Team’ is the right term for the company because I feel that Production and Maintenance are driving through change as a combined team effort to produce increased operation availability of better performing equipment. In time this will give us visibility into problems, repairs, equipment up-time and performance. Financial savings will come through efficient maintenance operations, reduced down time and reduced overheads. Within the short period of time that the new system has been operational it has prompted thinking about what can be done to improve communication – maybe we should install another terminal on the shop floor (as I have already mentioned this covers an area of 135,000 sq. ft., which necessitates quite a bit of walking to enter a work request), and a terminal will also be installed within the maintenance workshop to allow them to close down their own work orders and use CD catalogues for sourcing spares. The types of machinery on site complement themselves for condition based monitoring and this might be worth considering in the future. I have been looking at systems that can monitor various operational parameters, such as vibration, oils and lubricants, temperature, alignment, bearing characteristics etc., and these could all be linked back to the CMMS when an out-of-tolerance is detected. ONE YEAR ON The CMMS is now used 100% to record faults, costs, health and safety concerns for the building, plant and assets. It gives me a good tool for structuring reports – on costly repairs, on reliability and on plant which has undergone remedial refurbishment – to inform the establishment, to minimise production downtime, of routine maintenance at scheduled intervals. The staff in the department use the system to view history data and breakdowns, which has lead them to implementing design-out of failures. They have also been very active in logging all spares parts on site to establish a workable stores which identifies specific parts for critical machines. I myself have been talking to RS about their stock management system for the control of non-critical consumables. We are now in the process of targeting a selection of machines for monitoring and have undertaken training to carry out PLC programming and fault diagnostics in-house. We have investigated the possibility of the main production machines being fitted with modems and having direct dial-in link to the manufacturer for integration on machine stoppages. With what has been put in place the maintenance costs of our operation have been significantly reduced, and we are looking for further cost reductions in areas which have suffered from poor housekeeping and, due to production demand, very little maintenance. A targeted area for implementation this year is the creation of a single stores location, the movement of all current stocks and spares to this area, and the inclusion of this on the CMMS system. This will allow us to book out consumables to the asset and to know actual costs for all repairs and parts. I have set myself budgets, as targets to further reduce costs and this encompasses all maintenance on buildings, plant and grounds, along with service contracts to achieve the lowest costs compatible with no loss of service delivery. ‘CMMS gives me a good tool for structuring reports – on costly repairs, on reliability and on plant which has undergone remedial refurbishment – to inform the establishment, to minimise production downtime, of routine maintenance at scheduled intervals. ’
Click here to enquire..

Too small for a CMMS? Think again	Conference Communication
Too small for a CMMS? Think again Roger D. Evans President, Compliance Technologies, Woodstock, Georgia, USA Abstract Many smaller companies feel that, for them, the acquisition of maintenance management software would be overkill. The author explains (a) why he believes that nothing could be further from the truth and (b) the basic steps to be taken, by a small company, for the beneficial implementation of a CMMS. The problem Several years ago, at a small chemical plant, the manager was overheard complaining about the expenditures on spare parts for a process pump. ‘The pump costs only $18,000 brand new. How is it that we spent $14,500 in spare parts in one year? I have added up these costs several times. I kept thinking there must be a mistake; unfortunately, numbers don’t lie.’ ‘How much money did we really lose in production time?’ He questioned. ‘Why aren’t we smart enough to track equipment repair costs? We didn’t need to repair the pump. We needed to replace it. Downtime expenses, mechanics’ time, and spare parts combined, we have probably wasted $50,000.’ Why did this situation exist? The answer is simple. Many businesses have no way of tracking their maintenance activities. The ‘We are too small’ mentality ‘We don’t need maintenance software. It’s for big companies. We just don’t have the staff. We don’t have enough people to warrant the use of software. Maintenance software couldn’t possibly work here.’ In reality, even a one-person maintenance department can reap the benefits of maintenance management software. The same benefits realised by the maintenance crew in larger companies are there for smaller maintenance departments also. Smaller companies are typically forced to do more with less in nearly every area of their business. If they are not organised they will continue to work harder — not smarter. If the amount of time to administrate a repair or equipment failure can be cut in half, those unused resources are available for other tasks. Without software, nearly every time maintenance is carried out on a piece of equipment the small maintenance group will waste time trying to figure out the answers to recurrent questions such as – • Where did we buy that last spare part? • How much did we pay? • Do we have a warranty for this equipment? • Who was the salesperson we talked to? • What was the phone number? • Do we have an open purchase order with the company? • How was the last part shipped? • What was the delivery time for the last one we ordered? The maintenance person will probably get on the phone to accounting or other departments and ask them to research their records for the information. Again, more wasted time. Even with the most economical maintenance software package, most of this information can be right at your fingertips. Another important issue to consider is the amount of information that can leave the company when a key maintenance employee leaves. Years of critical technical information can be lost the moment the employee walks out the door. Implementation failure syndrome *‘Implementing maintenance software is easy; I’ve done it six or seven times so far.’* It is because of these failures that some smaller companies decide against buying maintenance software. Some studies indicate that, in some industries, maintenance software implementation failure rates are as high as 70 percent. It is not unusual to find a company that owns several different maintenance software products. Although software is usually the first point of blame when an implementation fails, in most cases human beings are the real reason for the failure. Many consumers of maintenance software have been led to believe that the only way the software will ever work is to spend thousands of dollars on implementation services. But end users can implement the software themselves. In many cases, they will do a better job than the software vendor because they are more familiar with their own plant. Implementation basics Implementing maintenance software can be quite easy if the end user has patience. Users should expect to write work orders in four to six weeks after software installation; however, achieving an efficient, smooth-running operation may take eighteen months or more. Getting organised is the first step in getting ready to use maintenance management software. This process can be started before purchasing software. Name areas. The first thing to do is assign area names to the facility. This may be as simple as calling one area the manufacturing area, another the warehouse area, and so on. Consider breaking these into sub-areas. The manufacturing area may be broken down into materials, product pre-assembly, final assembly, painting, packaging, and so on. Think along the lines of how maintenance activities are handled currently. It should be easy to relate the maintenance performed to a specific area. Later, a report can be produced that can be sorted by area. As an example, a list of all breakdowns in the pre-assembly area within a specific date range may be useful to pinpoint problem equipment. The more areas that are defined, the better the level of detail for future reporting. Keep the list of areas in a spreadsheet or other document. It will be more than likely that the information can be imported into the maintenance software. Name equipment. Naming equipment is one of the most important steps to success. The naming scheme should support future growth as well as the way the current workforce recognises the equipment. Conventional schemes such as ‘P’ for pump and a three-digit number (P-101, P-10A, or P-10B) should be considered. Some companies embed an area designation into the name as well. If P-101 is located in the pre-assembly area, the pump name might be PA-P-101. It is important to provide a name or tag number for any piece of equipment in the facility that could ever be maintained. This should include office air handling equipment, company vehicles, water heaters, compressors, etc. Again, place the list of equipment in a spreadsheet or document. Identify nomenclature requirements. Equipment nomenclature can be defined as the information required for purchasing the equipment or part without the need for the owner’s manual or without contacting the supplier. Establishing equipment nomenclature can make the life of the maintenance technician significantly easier. Consider creating nomenclature templates for different equipment or part types. As an example, each time a motor coupling is purchased, the supplier needs specific information to ensure the correct coupling is provided. General nomenclature templates to consider are pumps, bearings, belts, motors, control valves, gear reducers, instrumentation devices (level, flow, temperature, etc), and compressors. There will be equipment or parts that are unique to a specific industry. Nomenclature is particularly important for unique items because the equipment or part may have to be manufactured. This information also can be imported into the software; however, consider placing the nomenclature into a document file. Corrective maintenance. Corrective maintenance vs preventive maintenance is an often-discussed topic. Generally, industry guidelines recommend that 80 percent of the work done in a facility should be preventive maintenance and 20 percent corrective or reactive. However, when you are implementing maintenance software, forget this advice. Wait until the basic infrastructure of maintenance is in place and working well before venturing into preventive maintenance percentages. Instead, concentrate on establishing a corporate culture that readily accepts the mandatory use of maintenance software. A rule established early in the transition from a manual system to software might be: Effective (date), all work performed by maintenance department employees will be recorded on Form (form name here). The information to be recorded, at a minimum, shall include – • Area of the repair • Number of the equipment repaired • Start time of the repair • End time of the repair • Parts and consumables used • Employees involved Software is not needed to establish this requirement. The use of the information is twofold. First, it creates the beginning of an equipment history for the plant. Second, it provides the foundation for the culture of recording maintenance activities within the department. One of the biggest factors in the failure of maintenance software is the lack of willingness on the part of maintenance personnel to provide critical information to establish maintenance histories. The paper work orders can be easily entered into the software with the ‘Open and Close a Work Order’ feature in most maintenance software products. Preventive maintenance. What about preventive maintenance? Start with ranking the plant’s equipment according to its degree of importance. Start slowly. Identify equipment items that are required for the plant to generate revenue. Review the manufacturers’ recommended maintenance for the equipment. Then blend common sense from your maintenance experience with the maintenance the manufacturer is recommending. Next, create a maintenance task specification that includes – • Who is performing the work: maintenance or subcontractor. • Permit required to perform maintenance (lockout/tagout, confined space permit, etc). • Special tools required (include personal protective equipment). • Spare parts required. • Special lubricant(s) required. • Estimate of man-hours needed. • Description of task (fully explain the sequence of steps to perform work). • Description of appropriate test or check to confirm equipment maintenance is complete. Place this information in a document file so that it can be imported into the maintenance software. Once maintenance task specifications have been created, review the man-hours required to complete the work. Look at the available manpower capacity in the maintenance department before scheduling the first preventive maintenance work order. It is a mistake to schedule more preventive work orders than the current manpower level can handle. This creates a lack of confidence in the system and, more importantly, demoralises the workforce. The sense of accomplishment is lost and the impression is created that the department is not performing the work. Maintenance tasks have to be scheduled at intervals that are physically achievable by the manpower available. As an example, do not schedule 20,000 hours of overhaul work if only 15,000 hours of manpower are available. Work orders are typically printed for one week of maintenance. Every effort should be made to adjust the schedules so that if the department gets behind, work orders already out on the floor are completed first. Keep on working Maintenance software implementation is a work in progress. It can be as simple as entering a small amount of information each day. Over time, the software gets populated. Some companies enter the information when confronted with the need to perform maintenance on a specific piece of equipment. Others elect to populate the software all at once. Any of these methods will work. The important issue is to develop a culture in which maintenance personnel want the system to succeed. This can be one of the biggest avenues to success. Maintenance tasks, new equipment, new staff, new technologies, etc., all play a role in how the maintenance software can be best used to alleviate downtime and maintain efficiency. Maintenance software has been around for decades. The price of computer hardware is at an all-time low. Low-end maintenance software packages can be purchased for about the same price as a well-equipped PC. The excuses not to implement maintenance software get fewer and fewer each day. Take the plunge. You’ll be glad you did. The author may be contacted at – Compliance Technologies Inc., 135 Mirramont Lake Dr., Suite #135, Woodstock, GA 30189, USA Tel: (770) 926-0737
Click here to enquire..

Equipment readiness and visibility using Honeycomb maps	Conference Communication
Equipment readiness and visibility using Honeycomb maps Abstract A new tool, MERIT, that is receiving wide interest both inside and outside the military community, is described. It provides managers with a ‘honeycomb’ visual representation of maintenance performance – along several different performance characteristics, and within several different data hierarchies – and links a visual display with underlying repair cycle detail, indicating equipment level and part number level status, information which can be used to follow-up and focus on particular problems, identify potential solutions, monitor ongoing efforts, and identify responsible persons or organisations. B James M Reeve William and Sara Clark Professor of Accounting and Business University of Tennessee, Knoxville TN, USA B Michael Williamson Deputy Director, Studies and Analysis Department, US Marine Corps Logistics Command, Albany GA, USA NBThe opinions expressed in this paper do not necessarily represent those of the United States Marine Corps. Introduction The US Marine Corps describes its mission as to be a ‘Total Force in Readiness’. The concept of ‘total force’ includes all the elements of an expeditionary force, including troops, support personnel, and equipment. The concept of ‘readiness’ is the availability of all elements at the required time and in the required place. Thus, when a tank goes down, the commander must know when it went down, why it went down, and when it will again be available. If the elements are not ready, then the mission is not ready. Not surprisingly, a culture that is built around these requirements must also manage a maintenance system for fast response. Recently, a new tool has been developed and deployed to assist the Marines in planning and evaluating equipment readiness. Termed MERIT (Marine Corps Readiness Information Tool), it is providing the Marines with near real-time readiness information to support strategic deployment, planning, and management of equipment maintenance status and underlying parts. Already, its results are receiving wide interest both inside and outside the military. For example, US Congressman Mutha, speaking in May 2003, stated that In this article we will describe MERIT and show how it was developed to support equipment readiness performance and repair cycle visibility. Honeycomb maps The tool is a graphical analysis layer based on a new technology, termed ‘honeycomb’” mapping. Honeycomb maps, based on the theoretical work of Ben Shneiderman[1] have been developed and marketed by the Hive Group (hivegroup.com). The map expresses an information hierarchy as a two-dimensional mapping (see Figure 1). The highest, ‘blue’, level of the tree hierarchy is expressed in the honeycomb as the outer dimension. Lower tree levels (‘red’ and ‘green’) within the tree hierarchy are expressed by the inner dimensions of the honeycomb. For example, one of the most celebrated uses of this technology is in the investment sphere (see smartmoney.com maps). The market universe is represented by the outer dimension (blue level) of the honeycomb, as shown in Figure 1. The highest inner dimension is the industry dimension (red level), while the smallest squares within this dimension represent the individual companies (green level). In addition, attributes at any level within the honeycomb can be expressed by the size, colour and location of the various honeycomb cells. In the investment example, location represents an attribute such as market capitalisation. For example, telecommunications has the highest market capitalisation of all industries, which is indicated by its upper-left position on the map. Likewise, one level down, the Walgreen company has the highest market capitalisation in the retail drug sector. Additionally, the area of the honeycomb cell represents a second attribute, such as trading volume (ie the larger the cell, the higher the volume). Finally, a third data attribute could be captured by the fill colour of the cells (not shown). In the investment sphere, the colour codes would indicate the change in market price over a designated period of time. Red cells would indicate price declines, yellow cells minimal price changes, and green cells price increases. Thus, the honeycomb map is able to convey both data hierarchy and up to three data attributes within a single plane view. In more sophisticated applications, the individual cells can be linked to underlying data to provide drill down capability from the individual data elements (cells) to underlying cause factors. Such a visual display can be applied to a wide variety of scenarios, as we will illustrate with MERIT ( an application developed via a collaboration between Concurrent Technologies Corporation (CTC), the US Marine Corps and the Defense Logistics Agency). MERIT’s honeycomb layer resides over existing data providing dynamic information on equipment readiness (performance) and underlying supply-chain status (visibility). The MERIT front screen is shown in Figure 2. The hierarchy tree for MERIT is built around equipment groups and types. The equipment groups include Marine expeditionary equipment such as radios, trucks, light armoured vehicles, and the like. These equipment groups (or Functional Areas, FA’s) are visible on the MERIT screen as the larger boxes in white outline. Each cell within the equipment group includes a particular piece of equipment. For example, the upper left-hand cell labelled A2171 is a vehicular radio set. This hierarchy configuration can be changed by selecting alternative hierarchy definitions on the tabbed drop-down list. That is, users can change the honeycomb hierarchy by changing the group from equipment functional definitions to alternatives, such as commodity codes. The cells of the honeycomb are sized and colour coded according to two attributes, or performance metrics, which can be defined from the ‘size’ and ‘colour’ drop- down menus at the top of the screen. MERIT uses the size and colour dimensions to capture readiness characteristics of their equipment according to three measures, the S-Rating, the R-Rating, and the MR-Rating. The S-rating measures the number of equipment items on-hand (both available and in maintenance) as a fraction of the planned authorisations. The R-rating, or maintenance rating, evaluates the amount of equipment that is combat ready as a fraction of the amount available. The MR-rating is a summary rating, the product of the S and R ratings. Thus, if a Marine Corps unit has 125 units, but only 100 are authorised, then the S-rating would be 125%, indicating that there are pieces of equipment in excess of that which was authorised. These excess units can be located by Marine organisational units through MERIT view changes. However, if 50 units were under repair, then the R-rating would only be 60% (75/125). The MR-rating would be the product of these measures, or 75%. The MR is the percent of combat ready pieces relative to authorisation. These measures are very consistent with the broader class of equipment availability measures suggested in the Total Productive Maintenance literature. These measures can now be incorporated into the size and colour characteristics of the honeycomb map. The Figure 2 view shows the honeycomb cell size to be the current S-rating and the colour to be the current MR-rating. The S, R, or MR ratings for different time periods can be selected from the drop-down list to prepare different honeycomb views. Functional Area (FA) cells are one level higher in the hierarchy. Each of these cells is positioned on the map according to equipment density. To illustrate, ‘FA-10 Radios’ has the highest density of equipment in the Marine Corps, and thus is displayed upper-left on the view. The colours designate the current MR-rating. Green colours indicate MR-ratings above 92%, yellow colours MR-ratings from 85-92%, red colours ratings below 85%. Among the radios the A1955 (radio terminal set) has an MR rating below 85%, which would indicate the need for additional management oversight. In this example, because the S Rating sets sizing the individual pieces of equipment are arrayed within their functional groups so that the highest S-rated equipment is positioned in the upper left-hand (green) corner and the lowest rated weapon systems in the lower right-hand (red) corner of the group box. Thus, among the radios, the A2171 radio has the largest cell area, which represents the highest S-Rating among the radios. A click on the box will show the actual S, R, and MR performance attributes over designated time periods. Thus, users have a quick visual display of the multiple dimensions of equipment availability. Naturally, the honeycomb can be filtered along any of the equipment or performance metric dimensions to focus on a particular class of equipment or readiness issue. While viewing the maps from an equipment perspective is important, MERIT also arrays the data from an organisational perspective, which allows analysts to quickly focus on organisations that are experiencing problems. In addition, the system archives historical S, R, and MR ratings by equipment type for control charting, so an analyst can click on an equipment cell and request the system to prepare a control chart of the maintenance performance for variable time frames such as the last 24 months. Figure 3 provides an example for a light tactical vehicle using assumed data. Over the last twelve weeks it has been experiencing declining MR performance. The resulting honeycomb maps have allowed Corps personnel to move energy and time away from data gathering, accumulation, and reporting; towards solving and preventing critical readiness problems. Force commanders are given clear visibility of readiness trends, revealing potential problems and associated causes. System structure and development methodology The honeycomb map by itself provides a multi-dimensional view of performance, but does not provide the management information needed to influence the causes of the performance. In order to accomplish this objective, supply chain variables must be connected to the honeycomb items (equipment). In this way, performance, underlying causes, and associated responses can be linked. MERIT accomplishes this linkage. MERIT draws data from the Marine’s maintenance, supply, logistics, and distribution systems as well as from third party suppliers and transportation systems, as shown in Figure 4. The technology behind MERIT is an open-source Java-based programming technique. The common delivery method is through a web browser using a Java Applet processed on the server and connected to a data source such as Oracle, XML or delimited text. The graphical results are embedded in HTML and displayed by the user’s web browser. The final product is a small, flexible, file that runs on virtually any platform and handles a large number of users simultaneously. The integration of these data sources and the adaptation to MERIT was done under a rapid application development (RAD) methodology. Under this methodology requirements identification, commercial off-the-shelf (COTS) integration and database integration were accomplished in a compressed time frame. Such a flexible, responsive, and entrepreneurial development approach was well suited to an environment characterised by rapid changes in IT policy combined with multiple system modernisation objectives. Figure 5 illustrates a rapid application development framework[2]. Traditional IT development approaches are similar to the rigid design and development process within the construction industry. Requirements are turned into a building through a slow sequential process. Such an approach is reasonable with a building, since the final product is costly to change or repair once built. However, this is not the case with software, which can be developed and designed iteratively. A working ‘bare bones’ version can be developed rapidly and then improved and tested under the strain of use. As the software is used another round of requirements and improvements is then initiated. RAD places working software in the hands of users much more rapidly than the traditional approach. More importantly, iterative design methods will provide more realistic and useful design inputs from users. The user’s imagination is able to function more creatively within the context of use, than in the context of planning (or hypothetical use). Thus, the cornerstone of a successful rapid development process is the production of a useful working application at the end of each development cycle. We were able to integrate the data into the first MERIT prototype in less than three months because business processes had been defined and requirements existed. Consistent with rapid prototyping protocol, the users were co-opted into the process as beta testers. A web site was established where users evaluated the new tool, responded to enhancements, and contributed ideas for new improvements. This process led to an evolving tool that provided more and more of what the users wanted. In the next section, we will see how weapon systems and programme managers, maintainers, and analysts can access detailed information to initiate readiness responses. Merit repair cycle visibility Figure 6 illustrates the system’s repair cycle management capability. This illustration simplifies the actual system, by reducing both the number of data elements and levels within the system. Assume a light armoured vehicle (LAV-25) is represented in the red zone on the honeycomb display. Assume the LAV-25 has 15 vehicles authorised and three are in the maintenance shop, yielding an MR-rating of 80% (12/15) as of September 15th. Thus, in this example the LAV-25 is in the ‘red zone’ and requires further analyst attention. The analyst can click on the LAV-25 cell of the honeycomb map and connect directly into the equipment maintenance system to evaluate the maintenance status of the LAV-25’s. The equipment level view shows the maintenance status for the three LAV-25’s under repair, as shown in Figure 5. Vehicle number V023 had a due date of September 15th, but is now expected to be available on September 30th, or 15 days after the original due date. Since this vehicle was not released from the shop on the scheduled due date, the MR-rating has dropped below 85%. At this point the analyst can evaluate maintenance either from an organisational perspective (left-hand branch) or from a specific repair order perspective (right-hand branch). The organisational perspective provides the analyst with the complete detail of repair work orders, their status, associated equipment, due dates, and late status for a particular organisational unit. Naturally, this information can be queried and sorted using simple database tools. The organisational view allows the analyst or commander to evaluate the readiness detail of the entire Marine Corps, if necessary down to the smallest unit by evaluating the complete repair status of all pieces of equipment assigned to the unit. This view will reveal the underlying causes for the readiness status for a particular unit. For example, the unit readiness analyst could evaluate all the equipment in maintenance and re-assign resources to accomplish the most important tasks or to reduce bottlenecks. Commanders at higher levels could evaluate maintenance conditions and redistribute resources across subordinate units to increase the overall readiness of the command. For example, the commander may be waiting on a different but specific part for trucks. By re-prioritising the parts that are actually possessed, the trucks may be repaired more quickly than if left as an unattended process. Alternatively, the analyst can drill into the repair order view, as shown in the right-hand pathway in Figure 5. This view explains why the LAV25-V023 is past due, by indicating the reason for the repair and the timeline status of the repair. Many delayed repairs are due to part shortages. In order to evaluate the part shortage the analyst needs to access the material supply system. MERIT makes this access transparent. The analyst can click on the ‘Shrt Parts’ label in either the organisational listing or repair order listing to move into the supply system. The supply system provides the analyst with the ordered part detail. This would include the part number, due date, status, expected ship date, and other pieces of relevant part number information. From this detail the analyst can determine if the expected equipment due date is reasonable. As can be seen in this example, there is a discrepancy. The short part is due on October 5th, while the equipment is expected to be repaired on a revised due date of September 30th (from the equipment maintenance screen). Thus, the analyst is now aware that the equipment will not likely be ready on September 30th, as in the revised plan, but will more likely be available some time after October 5, after the short part is shipped. The analyst can either change the equipment expected due date or attempt to accelerate the shipment of the short part. To facilitate the latter action, the system provides capabilities to support inquiries to the purchasing agent, or directly with the supplier. The part number can also link to transportation provider tracking details if the part has been shipped, so that exact arrival status can be determined. In addition, the part number inventory information can be accessed from the part number hot link. Exception-based feedback One of the objectives of near-real-time management support systems, such as MERIT, is to build in feedback loops so that exceptions can be identified and monitored without the need for excessive database search and query. MERIT accomplishes this objective in two ways. The first feature is a screening mechanism that allows the analyst to segment the full database into a MyMerit view. MyMerit filters the database so that organisational units or equipment items are isolated. This provides focused analyst control at appropriate levels of responsibility. This feature is like establishing and tracking a stock portfolio. The second feedback approach is an email alert feature. The user can configure the system to provide email alerts when performance ratings drop below or improve beyond established thresholds. Finally, because tools like MERIT draw data from many different transactional systems it highlights many disparities between and among them. Accordingly, many reconciliation features were included, such as the following frequent conditions: • parts are not on order in the supply system even though the maintenance system indicates they are; • parts have been delivered to the requestor’s facility even though his system indicates they are still due in. Features such as these can be used to actively manage the maintenance and supply system to overcome equipment repair delays. Conclusion MERIT is a new tool that is receiving wide interest both inside and outside the Department of Defense community. The tool provides managers at various levels with a ‘honeycomb’ visual representation of maintenance performance along a number of different performance characteristics, within a number of different data hierarchy alternatives. The tool links the visual display with underlying repair cycle detail, indicating equipment level and part number level status. Such information can be used to follow-up and focus on particular problems, identify potential solutions, monitor ongoing efforts, and identify responsible persons or organisations. REFERENCES 1. Shneiderman B, Tree visualization with treemaps: a 2-D space-filling approach, ACM Transactions on Graphics, Vol 11, pp 92-99, January 1992 2. McConnell S C, Rapid development: taming wild software schedules, http://www.credata.com/research/rad.html The authors may be contacted via – jreeve@utk.edu “....the MERIT system is an exciting new tool that is providing a great advancement in our ability to maintain the readiness of the Marine Corps. … This system provides a near real-time visual display of everything we need to carry out our mission, from tanks to spare parts."
Click here to enquire..

'/maintenance/work/order/tracking/software/ HYDRAULIC TANKS | '/maintenance/work/order/tracking/software/ CMMS FREE | '/maintenance/work/order/tracking/software/ ELECTRICAL ENGINEER | '/maintenance/work/order/tracking/software/ ENGINEERING COMPUTER | '/maintenance/work/order/tracking/software/ CONSULTANT MAINTENANCE | '/maintenance/work/order/tracking/software/ ASSET | '/maintenance/work/order/tracking/software/ PROJECT MANAGEMENT CONSULTANTS | '/maintenance/work/order/tracking/software/ CHANGE MANAGEMENT CONSULTANTS | '/maintenance/work/order/tracking/software/ POWER PLANT ENGINEERING | '/maintenance/work/order/tracking/software/ EAM SYSTEM | '/maintenance/work/order/tracking/software/ PC MAINTENANCE | '/maintenance/work/order/tracking/software/ BAR ASSET MANAGEMENT | '/maintenance/work/order/tracking/software/ EAM | '/maintenance/work/order/tracking/software/ LAWN SERVICE | '/maintenance/work/order/tracking/software/ CONSULTANTS | '/maintenance/work/order/tracking/software/ INSPECTION MAINTENANCE | '/maintenance/work/order/tracking/software/ FLEET PREVENTIVE MAINTENANCE | '/maintenance/work/order/tracking/software/ HYDRAULIC OILS | '/maintenance/work/order/tracking/software/ BUILDING MAINTENANCE JOBS | '/maintenance/work/order/tracking/software/ MECHANICAL ENGINEERING |