Good afternoon. Here is a question that is currently plaguing us with regards to planning and scheduling.
History: We began planning & scheduling in June 06. Talk about culture shock. Before that 100% of all work was simply fire fighting. We are currently planning & scheduling about 10% to 16% of all work. Our planning and scheduling team consists of the Maintenance Supervisor, Plant Engineer, Plant Process Engineer (me), General Manager, Operations Manager (production) and my planner & scheduler. A very well rounded team that meets twice a week (Tue & Thru mornings).
Current problem: As the days pass by, more and more work is "falling behind". Basically they old fire fighting ways are pushing the planned and scheduled work back further and further. At one point when the work got months behind, the team thought it best to move all the work that was in the past forward in time. I was against this because I was afraid that this would become the solution and eventually there would be more work planned and scheduled in one day then I have maintenance personal to do. Well it's months later and guess what... the current planned and scheduled work is being slotted in the same time periods that the old work ended up in. The reality for us is that the work that is behind us, probably will not get done for the simple fact of it being "behind" us.
Does anyone have any thoughts as how to handle this backlog? Is backlog the right term? It goes against my gut feelings, but should we continue to grab all of the work that is in the past and move it forward in time, hoping that it will get done?

This message has been edited. Last edited by: Paul1559,

Paul,
This is a common problem when moving from a fire fighting mode. Usually the problem is that people have got used to instant gratification, and work gets done on the basis of who shouts loudest.
The solution lies in having
1. A proper work prioritization system with clear written definitions of high, medium and low priorities; If you find that High priority work forms more than 5-8% of total and/or medium priority work exceed 25% of total, check you priority definitions. Proper rescheduling is only possible when we have a pool of 60-70% low priority work
2. Enforcing these priorities rigorously
3. Having enough capacity to liquidate the average workload
4. Planned work must get priority, unless displaced by prior agreement, by critical breakdown or other high priority work
5. Have a process in place to eliminate defects progressively

See chapters 25-Workload Management and 27-Workflow Management of the book 100 Years in Maintenance & Reliability, details in footer, for some illustrative stories that could help you further.

Paul,

You must have a system that brings these jobs forward otherwise the only way they can be done is reactively by displacing work you have already scheduled for that day. However, this does not mean you just bring them all forward into the current week's schedule - they won't all fit. As Vee says you need a prioritisation system to assist.

We use SAP and utilise the Revision field to rank jobs for the schedule.
Revision
W00 Standing orders considered for every schedule
W01, W02 to W52 Correspond to calendar week. Assigned to jobs which MUST be actioned in that week
WR1, WR2 to WR5 Ranked jobs from WR1 - 'Really like to do' down to WR5 - 'Do if you have nothing better to do'

Note: For a job to be MUST DO there has to be some undesirable consequence if the job were not to be actioned in the allocated week's schedule. If there is no undesirable consequence then it is probably a WR1 job. A clear definition is required for MUST Do jobs and it this needs to be rigorously applied.

SAP Capacity Planning transaction is then used to dispatch jobs for the weeks schedule. First the W00 standing orders and the Must Do jobs are dispatched. We then review the % labour allocation and compare it with the maximum limit we have set for Must Do jobs. This limit will vary from site to site depending on the amount of break-in work that the site can expect. We have three sites using this system and the most reactive sets a limit of 50% whilst the least reactive uses 75%. If this limit is exceeded then we believe that the schedule as is will be unachievable and follow these three steps
1) Check that all the Must Do jobs really Must Do?
2) If still overloaded then can we schedule overtime to achieve the schedule?
3) If still overloaded then we need to engage additional labour to achieve the schedule

With the Must Do jobs now at a manageable level we then dispatch the ranked jobs up to 100% of the labour available.

Now that we have a schedule we then need to measure our performance. For each completed week we measure
A) The total hours worked
B) How many of these hours were worked on scheduled jobs?
C) How many hours were worked on true break-into-schedule jobs? - We don't like to work on these jobs but it is necessary. A good reliability improvement program will work to reduce this.
D) How many hours were worked on other jobs D=A-(B+C) - Strong management support is required to minimise this.

We present B, C & D by week as a stacked column chart to illustrate our improvement or otherwise.

Paul,
We were in a similar place not long ago. The "Urgent Work" gives instant gratification to Production and the Tech's like the Cowboy/Superhero attaboys that follow. We are now averaging less than 20% Urgent work...and not just on paper either. It took a real culture change to get there. Some things that helped were:
1) We required the operators to be operators... not just button pushers. Many of our floor calls were operator adjustments-no tools required. Our mechanics were the "superoperators". Procedures, machine centerlining and training sessions (often taught by mechanics) corrected this over time.
2) A formal work order request system allowed production an avenue to get their wishes done. The approved WOR were planned/scheduled then completed. We also provide requesters with weekly progress report of the status of their WORs.
3) During the weaning years, we rotated techs thru as "troubleshooters" to respond the the floor calls/urgent work. The remaining techs focused on planned work.
4) Good planned jobs made the techs more efficient. Once the planners got the hang of it the kitted parts and clear work orders cut down on the non-productive "research" time for the guys and upped the wrench time.
5) I would recommend posting the weekly schedule. This lets the crew know what's on the agenda and also lets Production know what they are pulling the techs away from...we are not waiting in the shop for the next call.
6) Large, time consuming jobs suffered most when we were reactive. "Why should I start this, I'll just get called away?" was the attitude. Breaking them up into bit-sized jobs < 2 hrs when possible helped get them completed.

One thing we didn't do well was developing a library of planned jobs. Now rather than improving each planned job as it re-occurs, we plan (steps, parts, time estimates) the same job over and over again.
As previously posted, the backlog is what we base our labor on. We try to keep a "Ready Work" backlog at 2-4 weeks. More than that calls for overtime. A sustained trend will call for staff adjustments. Too small of a backlog is also unhealthy- as long as the backlog consists mainly of work that is not critical to reliabilty. It should be kept as fill-in work to keep the techs productive.
We also use a priority system to move WOs along. We recently started to ratchet up priority on old low priority WOs to get them done.
Good luck!