<!– –>

Who’s really taking the planning decisions in your plant? The planning department or the shop floor?

In an ideal world, the planning department would plan and the shop floor would execute those plans.

The reality – as we all know – is quite different.  Schedules are rarely executed exactly as the planning department envisaged because stuff happens. Machines breakdown, employees call in sick, the required material isn’t available or turns out to be damaged, and so on and so forth.

So what should happen when the planning department isn’t around to reschedule? Which decisions should the shop floor be allowed to make? 

Here things get hairy. While shift leaders would obviously prefer full control over the schedule, they lack certain critical insights. They have no idea how their decisions will affect overall KPIs, and they lack the planning department’s soft knowledge of how a particular customer is likely to react to a delay.

Here’s a real life example of how the tension between planning and execution was resolved in a metals manufacturer.

The company concerned was struggling to improve its delivery performance. It had a scheduling solution and a manufacturing execution system (MES), and the workflow went something like this. During the day, the planning department created a schedule for the next shift. This schedule was then exported to the MES at the end of the day, in time for the nightly production run.

Every day the planning department created a great schedule (as measured by the relevant KPIs) and every night the shift leader ‘modified’ the planning department’s schedule: New tasks were added, and scheduled tasks were delayed, skipped or pulled forward.

These ‘modifications’ caused significant problems, particularly when:

• Urgent tasks were delayed to enable ‘better’ sequences
• Sequences were changed without considering the impact on downstream operations
• New tasks absorbed capacity reserved for scheduled tasks, thereby causing them to be late

 

The shop floor clearly didn’t trust the planning department’s schedules. This was hardly surprising as the schedules regularly ignored production rules, and were based on inaccurate estimates of production times.

It was obvious what had to be done. I added the missing rules and tuned the production parameters to improve the accuracy of the predicted production times. When I presented the results, both the planning and shop floor departments agreed on the following:

• The shift leader would check the feasibility of the schedule before releasing it to the shop floor.
• The shop floor wouldn’t push tasks into the future or add new ones. They could only skip tasks or pull tasks that were scheduled for the next shift forward. Whenever they deviated from the schedule, they would record their reasons for doing so in a log.
• Every morning the planning department would analyze the log entries and make the necessary adjustments to the production parameters.

 

These rules put the planning department in sole charge of all planning decisions. The shop floor’s responsibility was simply to execute those decisions.

The result? A significant improvement in delivery performance, and improved cooperation between the planning department and the shop floor.

To return to my question about which decisions shift leaders should be allowed to make.

All decisions that relate to the future should be taken by the planning department, as they have the necessary insight into how such decisions affect the supply chain.

All decisions that relate to the present should be taken by the shop floor. They should be allowed to skip a scheduled task (for example because the required material isn’t available) or bring a task forward. They should not be able to push tasks into the future or add new tasks.

Here then are my three tips for closing the gap between scheduling KPIs and real-world results:

(1) Define responsibilities clearly

(2) Measure the shop floor’s adherence to schedules, based on the relevant KPIs

(3) Respond swiftly to deviations from schedules by making the necessary adjustments to planning parameters

Do you agree with my conclusions, or has your experience been quite different? Let’s discuss.