Sample Maintenance Audit

                        Widget (International) Ltd

Report Presentation

(Click on a subject heading for navigation)

Section 1.          Site Operations

Section2.          Maintenance  Organisation

Section 3.         Operational Management Perceptions

Section 4.         Maintenance Policy and Strategy

Section 5.         Maintenance Planning and Control Systems

Section 6.         Compliance With Safety Regulations

Section 7.         Preventive Maintenance Activity

Section 8.        Performance Management

Section 9.        "People" Systems

Section 10.      Contractor Control

Section 11.       Stores Systems

Section 12.       Best Practice Benchmarks

Section 13.      Observations and Conclusions

Section 14.       Opportunities for Improvement

This report covers the results of an audit of the maintenance function at Widget (International) Ltd by Frank Chambers of PRIME on 12th December 2002. It is intended that the report will provide Widget with an insight of their maintenance operations such that it can be used to support a viable performance improvement programme.

1. Widget’s Operations

Widget is a US-owned manufacturer of engineering parts, and the largest manufacturer of fasteners for complex systems in Europe. The Birtley plant manufactures custom-designed parts in a wide range of sizes and arrangements, principally for use in the chemicals, process and aeronautical industries. The Birtley facility operates on a 3-shift, 5-day basis covering a 120-hour working week.                 {Top of Page]

2. Widget’s Maintenance Organisation

The Widget internal engineering team is managed by the Production Engineer who reports through the production line management system to the Operations Manager (the Production Engineer spends less than 50% of his time on maintenance management activity). There are a total of 9 craftsmen covering the production shifts as below :

Contractors are used to carry out servicing on site utilities and building infrastructure, coded welding support and for providing cover for craftsmen when required.                          {Top of Page]

3. Senior Management Perceptions of Maintenance

The improvement priority for the Maintenance Department is to improve the operational availability of the manufacturing process lines in order to strategically position Widget to be able to take up additional production activity without capital expenditure, or to reduce operational hours of selected production cells. Additionally, any possible improvement in the direct cost of maintenance, and in the impact of maintenance on product quality, would be welcomed.

Production management perceive the current maintenance operation to be reactive in nature, with less than satisfactory levels of preventive maintenance, and with a low level of effort directed to downtime reduction. The level of craft skills, application of engineering standards and the response to breakdowns is considered to be good, as is the level of safe working of the craft team. Management of engineering spares is perceived to be poor.

The maintenance department is perceived to be insufficiently integrated into the manufacturing operation. Overall, the ‘Value-for-Money’ rating of the delivered maintenance service is considered to be average.  {Top of Page]

4. Widget’s Maintenance Policy and Strategies

There is no written policy for the maintenance department. The maintenance strategies that are in place comprise a major service on key machines, and on statutory requirements.

Maintenance is not fully integrated into manufacturing management operations. A formal Kaisen system is in place for generating continuous improvement in manufacturing, including maintenance operations where appropriate. The differentiated maintenance roles and responsibilities for the Production Manager and the Production Engineer are not clear.

The effectiveness of the maintenance operation has not been benchmarked against other similar companies, either within the Widget group of companies or in a wider context.                      {Top of Page]

5. Widget’s Maintenance Planning and Control Systems.

The system for requesting maintenance services is by raising a maintenance work request form although this is often not followed, leading to a predominance of verbally requested work.

The only work which receives any forethought and planning is for the preventive maintenance work. All of the remaining (85%) of maintenance activity is of an emergency breakdown nature, precipitated by equipment failures that halt the production process.

50% of plant is subject to some form of preventive activity, this being solely for a major service on key equipment, which has been in place for about 18 months.

No records are retained in any formal way for maintenance history purposes although it is possible that some information could be retrieved laboriously from paper records.

There is no form of CMMS (Computerised Maintenance Management System) in place, although an evaluation of the Shire "Frontline" system is currently being carried out.

Maintenance work procedures tend to be unwritten, relying on an individual craftsman’s experience. Although up-to-date manuals exist for most equipment they tend not to be used by craftsmen.

There is a clear identification system for all equipment, used by both production and maintenance personnel, with commonly used and understood I/D numbers.  {Top of Page]

6. Compliance with Safety Regulations

The occurrence of LTAs in the maintenance workforce is very low, with no accidents reported in the past several years (Best Practice levels would be around 1 LTA in more than 0.5 million man-hours). Minor accident levels are also fairly low, with a level of about 1 minor accident in 7000 man-hours. Compliance with the most applicable legislation suggests a generally satisfactory level of compliance although there are a number of areas where compliance appears to warrant more detailed auditing, namely :

· The absence of recording of maintenance work (this is an implied requirement under H&SAW Act and PUWER Regs).

· Compliance with the requirements of ACOP L8 for Legionella appears to be incomplete, in that there appears to be a lack of detailed layouts and risk assessments for hot/cold water distribution systems, and no regular monitoring of hot water temperature profiles.

· There is no copy of the Written Scheme for Examinations of Pressure Systems available on site, and the Insurance Company responsible for inspections is not always informed when modifications are made to the system.

· There is no formal lockoff/tagging system for electrical isolations although the standard practice is to isolate and remove fuses. However, as the electrical distribution boards are not secured, this would not be seen to be adequate.

· There appears to be a lack of regular, formal testing of machine guards, required under PUWER Regulations.                                  {Top of Page]

7. Preventive Maintenance Activity

PPM programmes are applied to approximately 60% of the production equipment. The PPM routines, which were introduced in the past six months, are based around single major service inspections developed from maker’s recommendations. The major service inspections, which are scheduled on a calendar basis (ie they reflect theoretical, not actual, runhours), are totally intrusive and typically incur about 50 hours of equipment downtime to carry out. If the equipment is fully occupied, this will involve a downtime level of about 2.5%; if the equipment is only operational on 2 shifts, the overall downtime level will be 4%. No form of formal FMECA (Failure Mode, Effect and Criticality Assessment) has been used to assess the effectiveness and appropriateness of the PM programme.

A programme of equipment lubrication is included in the major service inspections and there are no interim lubrication schedules.

There is no application of CBM (Condition-Based Maintenance) or routine running equipment inspection, either ‘3-Sense Testing’ or supported by inspection tools.

No routine checks are carried out on equipment alarms and trips systems.

The PPM tasks are currently scheduled using a paper-based system although it is planned to move to computer-based scheduling in the future, which will also provide for scheduling by runhours as an alternative to calendar scheduling.

8. Performance Management

There are essentially no measures for maintenance performance in terms of its impact on production. Equipment downtime is included in the operator’s shift reports, but the data is not used for routine reporting and is not communicated to the maintenance team.

Because downtime information is not available to the Plant Engineer, there is no formal system for downtime reduction based on Root Cause Elimination.

The use of OEE (Overall Equipment Effectiveness) as a tool for measurement of all throughput losses has been evaluated in the past although no scheme for formalising OEE measurement across the plant has been considered.

There is no system for budgeting maintenance activity, or for reporting maintenance cost in a way that would support a cost reduction programme. However, year-on-year cost reductions have been achieved in the maintenance area.

No annual management objectives are set in the maintenance area.              {Top of Page]

9. "People" Systems

Craftsmen have a relatively high level of empowerment to carry out work unsupervised. However, there is little cross-skilling between mechanical and electrical crafts, partly as a result of the traditional demarcation lines drawn by the craft unions to which all the craftsmen belong. Engineering staff carry out almost all maintenance work - production operators are involved in virtually no maintenance-related activities, and in fact are not even effective in regular machine cleaning.

Relationships between management and the maintenance workforce are good, although it is difficult to introduce effective change due to the entrenched Trade Union culture.

Overtime levels are high, at 15% of worked hours.

The training level is less than 1.5% of the maintenance budget.       {Top of Page]

10. Contractor Controls

There is no system of contractor evaluation prior to selection, or of regular auditing of performance. Simple ‘Rules for Contractors’ are in place and are applied whenever a contractor presents himself at the plant, but are often not enforced. Contract companies can supply any craftsmen, with no controls in place for confirming their individual competence.

No partnership or single-supplier arrangements have been investigated. Service contracts are renegotiated formally when they are due for renewal.          {Top of Page]

11. Stores Systems

The engineering store is unmanned, with free access to all. The stock, which is not insubstantial, is unmanaged in that there are no records of inventory quantity or value, and no recording of issues and receipts. It was reported that a considerable amount of craftsman time is lost searching for materials.               {Top of Page]

12. ‘Best Practice’ Benchmarks

Because of the difficulty in making precise comparisons between differing manufacturing operations it is not advisable to draw distinct conclusions on the basis of maintenance ‘Best Practice’ metrics comparisons alone, nor should they be viewed as being entirely indicative of the situation at Widget in view of the relatively small scale of the maintenance operation. However, they do provide insights into the areas available for improvement and can indicate the potential scale of any improvement opportunities.

The metrics which have been compiled for Widget’s operations based on the information provided to PRIME are presented the separate Metrics Report and Profit Datasheet documents. General conclusions that can be drawn from the comparisons of the key performance indicators are:

· The results indicate that the cost of maintenance at Widget as a proportion of all manufacturing costs is higher than in the best companies. This adverse comparison also applies to the ratio of maintenance cost to the replacement value of the plant and equipment. The amount of replacement value and maintenance cost borne by each craftsman is lower than Best Practice levels. These comparisons would suggest that there are opportunities to improve both the effectiveness of the maintenance operation and the internal efficiency of the maintenance team, although there may well be organisational constraints that will limit the degree to which cost reductions can be achieved without affecting equipment support levels.

· There are no precise measurements available for equipment downtime due to engineering although anecdotal evidence based on sampling would indicate that downtime is no lower than 8%. Over 2% downtime is incurred from the existing PPM programme alone.

· The backlog level is low. This usually leads to low efficiency in the craftsmen team because of the inability to cope with low activity levels, although this is probably not significant at Widget at present because of the relatively low staffing levels. A low backlog level also tends to drive toward a reactive maintenance culture in both maintenance and production, and this is true in Widget. However, the lack of a system for managing work requests probably works against jobs being raised at the earliest time, and the ‘real’ backlog may well be greater than 1 week. About 2 to 3 weeks of backlog would be usual in an effective and well-regulated maintenance environment

· The indicators showing ratios of direct employees to the numbers of indirects, supervisors and planners are unlikely to be appropriate for comparison at Widget due to the relatively small numbers of people involved.

· The low level of expenditure on training is probably another element in the relatively low effectiveness level.

· The reasonable level of PPM coverage at 60% of all equipment is not consistent with the high downtime levels and the high level of emergency breakdown work. This suggests that, either the PPM programme is not optimised and has not resulted in improvement in reliability, or that the craftsmen are inactive at times when there are no breakdowns, or that both situations apply.                  {Top of Page]

13. Observations and Conclusions

There are a number of important conclusions that can be drawn from the audit.

· There are two ways to minimise equipment downtime due to in-service failures, firstly to improve plant reliability by eliminating repetitive failure causes, and secondly to minimise the duration of repair intervention events. In the past Widget have concentrated almost exclusively on the latter approach, resulting a well-received repair culture. However, there are no long-term reliability improvement programmes in place, and therefore continuous management effort is required to maintain any gains in performance that have been achieved by the fast response approach alone. There is a need to review the strategies for maintenance to ensure that they are focussed on long term, self-sustainable performance improvement that does not need to be management-enforced.

· Although there are measurement systems in place for downtime, this data is not available to maintenance and is not structured to provide the information they would need to support an improvement culture in maintenance. It has been subjectively assessed that OEE levels, although not measured, are probably less than 50% (this value is consistent with what would normally be the starting position for companies which had not used historically used OEE measurement). In this case it is likely that downtime resulting from in-service failure of plant is likely to be no less than 8-10% (in line with the anecdotal evidence indicating that it is no less than 8%).

· In conjunction with the lack of engineering downtime data, the lack of quantitative maintenance history from maintenance work records prevents the practical application of an objective Root Cause Elimination programme.

· Based on some assumptions made against the estimated OEE values, the total cost of maintenance, including both direct and indirect costs, would appear to be in the order of £1.3M (a cost escalation factor of 3.1 on direct cost against Best Practice norms). This would certainly suggest that there is considerable opportunity for performance improvement.

· The design of the preventive maintenance routines are probably not optimised in that they have not been based on any objective assessment of failure causes, have essentially used maker’s recommended repair tasks, and have been developed around intrusive maintenance techniques.

· A simplistic cost/benefit assessment using the top level KPIs in the metrics report would strongly suggest that significant performance improvement should be achievable without any overall increase in maintenance cost. The KPIs would also suggest that cost reduction should be possible, but this needs to be assessed in the context of the lowest practical manning levels that would be necessary to ensure satisfactory shift-by-shift coverage, and the special skill sets required at Widget.  {Top of Page]

As described in Section 13, the results from the audit suggest that there are a number of opportunities to improve the effectiveness of the maintenance service in Widget. The main recommendations arising from the audit are as follows:

· Establish a simple, clear policy/strategy for the maintenance function, agreed by all of its ‘customers’, and based upon the current business drivers. (The ‘customers’ would normally comprise Production, Finance, Safety and Quality departments). This may appear to be a somewhat pedantic proposal, but experience indicates that it is a valuable precursor to establishing the foundations for an effective maintenance system. The responsibilities and objectives of all of the people involved in the management of the maintenance function would be defined in such a policy document.

· Develop a structure to drive continuous improvement in the Maintenance area, primarily in the areas of availability and cost. This would require the establishment of performance standards, measurements of performance, and a system for initiating and executing improvement activity through root cause elimination and the removal of avoidable work. Such a closed loop approach to improvement lends itself to the introduction of objectives and targets for improvement, and could be carried forward into processes for lifecycle costing for selection and operation of new equipment.

· Carry out FMECA on all production-critical and safety-critical equipment to generate the optimum maintenance strategies, concentrating on CBM (Condition-Based Maintenance) wherever appropriate. Experience across all industries indicates that CBM is almost always a more effective basis for preventive maintenance than fixed-interval maintenance in those situations where it is technically appropriate. There are simple tools for carrying out effective FMECA that would be applicable at Widget and it is recommended that they be initially used on one critical piece of equipment for evaluation.

· Complete the planned evaluation of a CMMS system. Based on the low cost of modern systems it is likely that an entry-level system would prove to be justified in Widget’s operational environment (PRIME can supply a simple evaluation tool to assist with such an evaluation). The evaluation should extend beyond the scheduling of PPMs to the broader need for relevant data to support the improvement programme. Basic requirements for this would be the provision of a formal work request system, the visibility of all maintenance work (current and in history, ideally including materials and repair times/costs), and facilities to associate and interrogate relevant failure types/failure causes with breakdown events.

· Introduce a formal downtime measurement and reporting system, including categories of downtime events to improve the visibility of different types of equipment failure. Set up a analysis system to support simple evaluation of downtime, ideally with a cost base as well as time (PRIME can supply such a system free of charge to client companies).

· Consider further the value of an OEE-based programme for achieving overall production loss reduction. Indications at Widget would suggest that the current level of OEE is probably around 50%. In complex batch plant, a ‘Best Practice’ OEE level would be around 85%. The theoretical benefit to production recoveries by improving OEE to around ‘Best Practice’ levels would, therefore, be in the order of £2.1M per annum, with the maintenance contribution probably in the order of £950k pa (the term ‘theoretical benefit’ refers to the assumption that any additional output accruing from OEE improvement could be converted into sales).

· Introduce a system for condition-based inspection of all plant based on "3-Sense Testing" involving both production operators and craftsmen, as the starting point for a CBM approach to maintenance. It is very likely that the FMECA approach suggested for determining the optimum PPM regime would result in the present intrusive maintenance tasks being significantly replaced by CBM tasks, including "3-Sense tests" wherever relevant.

· Evaluate the use of acoustic emission measurement as the front-line tool for the condition assessment rotating equipment, again as indicated in the FMECA studies suggested. In Widget’s type of operation this is probably a more effective tool than vibration analysis for assessing bearing failure, and is non-intrusive to production. Simple methods and equipment are available that could be used directly out by either operators or craftsmen.

· Develop structured methods to involve production operators more fully in the maintenance of their equipment. This will extend the principle of ownership to the group that is best positioned to monitor and care for the health of manufacturing equipment. Wide experience in industry has confirmed that the extent of equipment maloperation and abuse is inversely proportional to the extent of operator care, and it is usual for all low-level maintenance work to be done by operators, freeing up maintenance craftsmen to do higher level work. In the best companies, up to 30% of maintenance work might be carried out by operators.

· Introduce a system of budgeting and budget control for maintenance.

· Examine the benefits to be gained from extending cross-skilling in the craftsmen team. As the maintenance workforce have been with Widget for many years it is likely that they will have some difficulties in becoming confident on some of the recently acquired plant (eg the robotic assembly equipment), and training in diagnostic techniques may also be appropriate for both mechanical and electrical trades. In view of the entrenched Trade Union culture, a basic ‘cost/benefit’ analysis should be carried out to decide whether there is sufficient incentive to implement this aggressively, and if so, to what extent any additional cross-skilling should be implemented. If so, develop a strategy for implementing the changes indicated.

· Review the level of compliance with L8 ACOP for Legionella.

· Introduce lockoff/tagging systems for electrical isolation, and a programme of routine testing of machine guards (recommended on a shift or daily basis).

· Review the level of compliance with the Pressure Systems Regulations.  {Top of Page]