Support or straitjacket

How do you know if your site is operating well? The obvious way is to see how you match up against others. Probably the comparison that matters most is how well you stack up against your biggest competitor. If it is delivering faster, better, cheaper and more profitably than you, you've either got to at least draw level or give up now. But seeing how you compare with others in the same sector, or those using the same processes, or even how you fare against the best operations in the world can be enlightening, whether it spurs you to improve or to slit your throat. Accelerate, the European Regional Development Fund advisory body for West Midlands automotive suppliers, has been using a benchmarking process for 12 years to help these firms become more efficient. "Every company should be aware of how their performance compares with peer groups of similar size, type and products, enabling them to identify their strengths and weaknesses through key operational criteria, such as return on capital, gross value added, overall equipment efficiency, and so on," says Accelerate adviser Norman Taylor. "These results can then be used to establish business goals and future strategies." In Accelerate's experience, SMEs that have used benchmarking diagnostics as a basis for improvement have survived both the closure of MG Rover and overseas competition, and emerged as more robust and successful. It now intends to provide companies with online diagnostic benchmarking to guide specialist advice and training. This kind of thinking has launched a whole raft of benchmarking information from which individual companies derive KPIs (key performance indicators) to shape their own improvement activities. BERR and industry forums are major sources. Cranfield holds a massive database and is, indeed, the source of the benchmarking which is one of the most important elements in the judging of the Best Factory Awards. The EQFM's (European Foundation for Quality Management) benchmarking model is also widely used and respected. Probably one of the best known, however, is Oliver Wight's 'Class A' checklist, an icon in performance measurement. Hundreds of manufacturers, many of them global names, have bought into its approach of educating, coaching and mentoring people to sustained change and the ultimate goal of 'business excellence'. Today the checklist is in its sixth edition and aims to encompass the whole business, not just operational aspects. So there are now chapters dealing with people and teams as well as driving business improvement. Class A is awarded only to companies meeting a minimum performance standard of between 95 and 98 % measured against the checklist. "The measurements within Class A are a better ring-fence of a business than others - it doesn't allow people to escape," maintains Oliver Wight consultant Les Brookes. "If you fudge it in one area, it will pop out in another. To achieve it, we insist on three consecutive months with all of the key performance metrics achieving the minimum level. You have to have good processes to do that; you can't do it through energy alone. It demands that everyone works naturally through the process - it can't be pushed for the audit alone." Similarly, he believes that the integrated approach is a far stronger test of the overall organisational capability: "For instance, it is easy to achieve high levels of efficiency at the expense of schedule adherence: 'We can make anything anytime but we don't make it on time'. The integrated measurement is vital." Journey of pride There's no doubt that the Class A road is a hard one to travel and those who have got there are rightfully proud of their achievement. At Oliver Wight's regular Proven Path meetings, the air is thick with messianic fervour. I have even heard people ask "When did you get Class A?" in the same envious tones they would use about a Lottery win. Nonetheless, even in a standard as stringent as this, it is still possible to meet the KPIs without addressing some of the underlying inadequacies in the business, as the Ruabon site of Flexsys clearly testifies. Ruabon Works, one of the oldest chemical sites in the UK, employs about 180 people and manufactures a range of complex 'effect chemicals' for the rubber industry. After a major education and improvement programme, it was awarded Class A in planning and control. As a result, its supply chain operation was in excellent shape and the plant was hitting the majority of its KPIs, yet it still found that it couldn't consistently meet all of its production planning targets. The key measure for production against the monthly plan was consistently at or close to the right level. The day-to-day performance to plan was close, albeit with more 'noise' in the numbers. The weekly measure, however, was struggling to reach half the target. "At manufacturing level, all of the processes were in place and in theory everything was perfect," recalls engineering manager Roger Mason, "but it just didn't work and it was like pushing rope uphill every week." So he decided to look behind the performance numbers. "Almost always it was something to do with the plant breaking down. In an asset intensive industry without a huge variety of SKUs or many manufacturing steps, asset reliability has a bigger impact than all the intricacies of how you route things through the factory." Ruabon could hit the key monthly measure because demand was largely below overall capacity so, if there was a breakdown, they could just 'turn up' the plant to make up the shortfall. "But the one we kept missing was the weekly, because we couldn't predict where we would break down and there wasn't enough time to get it mended and catch back up again. So all of our behaviours became focused on very reactive repairs and every day was an emergency to try and hit the green flag. Yet we were still hitting the KPIs that the business was telling us were important." This is a clear example of where a KPI can become a straitjacket rather than a support. It can narrow a manager's focus to delivering the numbers rather than taking a wider view of what the business needs. Mason realised that without some structured way of managing reliability, Ruabon would never improve. To kick off that structured process, he called in engineering consultants MCP to perform their AMIS (Asset Management Improvement Service) audit. Yet another benchmark, of course, but one designed to address a specific problem preventing the plant from meeting its top-level management plan. With a score of 38%, MCP's Richard Jones described the plant as 'World Class Reactive'. Over the next 12 months, Ruabon worked with MCP to redesign the entire site maintenance process to one centred on asset care. The project team, which included operating and support representatives, actually used the principles of demand and supply management learned from Oliver Wight in developing a new work order management process. On its first audit, Ruabon's score rose to a creditable 60%. Ruabon wanted to reach world class levels of 75% plus. To do so, however, it first had to address some fundamental issues beyond the reach of normal benchmarking. Culture change During the first year, Ruabon had questioned if the achievements were 'in the bones' of their team. The change from reactive work to structured schedules seemed difficult for many people on site. There were no overtime heroes bringing salvation from breakdowns; such episodes were now classed as failures. To some, it felt like their core competence was no longer valued. For such a significant change to work, these deep-rooted issues of individuals' behaviour and sense of self-worth had to be tackled. The management team introduced personal and team development programmes to look at behaviours and interaction. They established an open culture of individual awareness, using one-to-one feedback as an everyday activity to be sought rather than feared. "We set out to help people believe that they were just as important working to prevent things breaking," recalls Mason. "Finding KPIs to demonstrate things aren't going wrong is quite hard - like the safety manager who never knows how many accidents he's saved, just the times when someone got hurt." So instead they put KPIs on the asset care inputs like root cause analysis and preventive work, and backed off measuring time spent in fixing breakdowns. Some of the new KPIs were ingenious. For example, the computer was set to show how often someone collected something from the stores instead of having it delivered; a clear indication of unplanned rather than planned work. "It indicated how well people were thinking about their jobs and it let us measure that behaviour without it being too intrusive." Mason is adamant that if you don't think about behaviours, KPIs will create them. "If our people don't believe we want to run the factory in a way they value, they will do what we tell them only as long as we are looking at them. Cultures are defined by what people do when no-one is looking at them. We wanted everyone to act honestly and we asked people to challenge us if our behaviour wasn't consistent with what we were telling them we wanted to do." Mason draws a sharp line between what KPIs can and can't do. They are necessary to measure the system's capability and to tell you what is actually happening. "But if they are then used to set targets that have no relationship to that capability, all it will do is create resentment and stimulate people to find clever ways to con it. The system will deliver what it will deliver. Rather than issue a target from on high, you have to go back and find out what is creating this output. It could be behaviour of people, performance of the machines or, in most cases, a bit of both. "You should use KPIs to drive your improvement rather than control your people. The command and control approach just plain doesn't work - when you stop looking they'll go back to doing it the old way." Assuming that the KPIs are well chosen and describe what a business has to deliver, Mason believes they are vital. But he's not so sure of the truth of the old adage that says if it can't be measured, it can't be managed. "Human beings are more complex and maybe we should trust our intuition more." Two-pronged attack After four years, Ruabon has moved from 38% to 84% on the AMIS audit, putting it close to the highest score in its sector. Of even more importance to the site as a whole, the weekly production planning performance measure has doubled and it is regularly achieving Class A performance of 95%. Does it matter that Ruabon is measuring itself against two benchmarks? "You have to have a sense of what you are trying to achieve. Oliver Wight is not aimed at asset reliability, AMIS is. We wouldn't, however, have come to this issue if the Oliver Wight measurements hadn't pointed out, that with all of the changes in our supply chain and in integrated manufacture and inventory management, we were being let down at core level by reliability. "You have to look at the whole suite of KPIs, see what's good but recognise the issues that won't go away. You then have to look at new numbers to discover what is happening and you may well need another set of KPIs to measure it. So you can use a whole mixture and you should be quite ready to jump ship part way along. Even if there is religious belief in a certain measure, it's sensible to drop it if it doesn't support what we are trying to achieve at the moment. "After all, the difference between being in a rut or a grave is only the dimension."