A large pulp and paper mill complex in the US northwest was performing in the bottom quartile when bench-marked against other mills. Further benchmarking of the mill's process performance in Maintenance and MRO Materials Management showed they were similarly low ranked. One process that was in need of improvement was Maintenance Work Management. Spending on maintenance was on the order of 25% of operating costs - that presented at cost savings opportunity of 20% if practises would improve. At the time improvements began backlog was measured only in terms of work orders outstanding and incomplete. Over 850 of the work orders (nearly 75% of them) had top (safety related) priorities associated with them. It was broadly accepted that assigning a top priority was the only way to get any job done. That number was growing daily and no one was really confident that all the work was properly recorded. The work order system (they were using the maintenance module in a popular enterprise resource planning system) was underutilized and avoided by many. Maintainers were almost exclusively tied up in repair work dealing with breakdowns just to keep the mill operating. The maintenance department was really just a "repair department".
Once we had a full assessment of the situation our first step in improving the situation was to get it under control so that legitimate safety concerns could be addressed, backlog could be reduced and resources freed up for some proactive (preventive and predictive) work to slow the growth of repair backlog. All backlogged work orders were reviewed - about 10% were eliminated as duplicates, about 700 had no real safety related concerns and were downgraded and the rest were treated as truly urgent. With no formal job planning all jobs had their work durations and number of trades-persons estimated. Using those estimates the jobs were scheduled by priority, taking into account equipment criticality to production using only a spreadsheet. A graphic "calendar style" schedule was produced for the coming week for each shop. Given the high level of emergency work that still prevailed only 80% of the available trades-persons' time was scheduled to allow for those inevitable emergencies. At the end of the each week the designated planner (a newly created role) colored in the completed jobs so that work order completion could be observed graphically. The schedule for the following week was posted.
Completion of scheduled work for the week (i.e.: schedule compliance) was 50% in the first week. The backlog actually fell a bit. This started a trend - the trades-persons wanted to see their own shops have the best compliance so they went the extra mile to improve that compliance. They began to give feedback to the planners about job estimates so that they would be more accurate. Within a month, schedule compliance had climbed to over 80%. Backlog came down and now that estimates were available for all jobs we could see that backlog was approximately 3 weeks per trade - a healthy target.
As work order schedule compliance improved the production departments began to assign more realistic job priorities. The panic subsided and control was established. In this more stable environment other improvement initiatives like complete job planning, initiation of performance measures and reliability enhancement (initially using Root Cause Failure Analysis) could begin. The resources were now available to be proactive.
Begin work management improvement with simple broad steps - no need to be perfect.
The benefits of improved work management processes and practises with broad based acceptance across the organization are a reduction in the chaos associated with frequent breakdowns so that improvements focused on longer term issues like reliability can be addressed systematically and thoroughly.
