Trivia Question: Should You Reschedule Production Whenever Machines Go Down or Workers Call in Sick?
- Posted by altemir
- On 01/13/2019
- 0 Comments
We examine the circumstances under which production rescheduling can either be helpful or hurtful in driving on-time performance. The answer may surprise you.
I suspect that the majority of manufacturing and production control managers believe that their ERP shop floor calendars should ideally be updated whenever unplanned events occur in order to keep the production schedule accurate.
Here, “unplanned events” can refer to any of the following situations:
- Machine breakdowns
- Unexpected staff absences
- Material shortages
- Unplanned material scrap
- In-process engineering changes
These issues are all similar in that: 1) they were not anticipated in the manufacturing routings and calendars, and 2) they adversely affect the company’s ability to supply product on-time. It is worth noting, however, that none of the contingencies listed above reflect a change in customer demand. In that sense, these unplanned events are all load- and capacity-related supply-side issues.
I want to challenge the assertion that it is always appropriate to change production schedule dates (i.e., “reschedule”) whenever unplanned events occur. To do so, let’s consider a manufacturing operation for a part number that lies near the bottom of a multi-level bill of material. If, let’s say, a machine breaks down, it’s then clear that the breakdown will likely delay the completion of the operation and, if the operation lies on the critical path, the delivery of the final product may also be delayed as a result.
If we reschedule accordingly, then an entire string of scheduled due dates will move to the right to show the schedule impact of the breakdown. In this case, we can probably all agree that it’s useful to know how unplanned events affect estimated completion dates (ECDs).
So far so good.
However, the customer may not be agreeable to a later delivery and may insist that the original due date remain. In this case, it is useful for the manufacturer to know for each operation at every level of the bill of material that an operation is due, for example, on Monday but realistically won’t be completed until Thursday (i.e., it is projected to be three days late). We are then able to know that Thursday is the current ECD because we changed the production calendar to reflect the effects of the machine breakdown upon available capacity. But, in so doing, the Monday completion date is overwritten with the newly projected completion date, Thursday.
Therein lies the problem. If we operate with a single production schedule where we allow unplanned events to drive date rescheduling, then work priorities for all the downstream operations in the chain will be shifted. This is particularly problematic in job shop environments where many different manufacturing jobs may be competing for the same resource. The priorities for the impacted product will have been relaxed relative to the other competing jobs. Operators will now think that the operation is due on Thursday whereas the customer requirements suggest that a Monday due date is more appropriate. This date rescheduling diminishes the prospect of the factory making up time to put production back on schedule because the schedule priorities now tell them that a Thursday completion is OK, which the customer says is not.
The scenario just described leads us to the conclusion that sometimes it’s useful for the schedule to reflect the impacts of unplanned events upon the supply chain so we can know when to expect work to be realistically completed. However, we should also recognize that scheduled due dates based on customers’ requirements, this time where the impacts of unplanned events are excluded, are useful for maintaining customer-driven work priorities in the interest of meeting customer commitments.
Properly managed work priorities are one of the most important levers a manufacturer has to drive on-time performance. If our priorities change every time some eventuality occurs on the shop floor, then we can count on the production schedule becoming immersed in volatility and, ultimately, ineffective as a tool for on-time shop floor execution.
So What’s the Solution?
We have to ask ourselves what is the purpose of a production schedule? Is the production schedule supposed to drive the shop floor or vice versa?
We can look at this question in a variety of ways, but I think the most straightforward way is to say that, first and foremost, we want the schedule to drive shop floor activity in a way that promotes on-time completions that satisfy the customers. But we can also acknowledge that, at the same time, we want some kind of schedule feedback from the manufacturing activity that tells us when work is realistically projected to be completed. In essence, we want to be able to know that an operation is due on Monday but is projected to be completed on Thursday.
In short, sometimes we want the effects of unplanned events baked into our schedule dates and sometimes we don’t. We want it both ways!
The practicalities of managing ERP manufacturing scheduling in a way that maintains customer-driven due dates while at the same time providing reliable estimated completion dates (ECDs) is addressed
in my book, Lean MRP: Establishing a Manufacturing Pull System for Shop Floor Execution Using ERP or APS.
In it, I explain that two independent schedules need to be simultaneously managed by the system: A “BASELINE” schedule that is stable and focuses on communicating customer-driven priorities to drive the shop floor and a “PROJECTED” schedule that is dynamic and is periodically updated to provide reliable ECDs based on progress to date and available capacity.
The following three-minute video provides a short synopsis of the “Lean MRP” methodology, which can often be supported with your existing ERP system.