I have been involved in several projects and sales conversations where the requirement is to get to an optimal schedule. Upon probing, most people confirm that they want the mathematical best (AKA optimal) schedule that achieves their desired objectives. Some of these objectives include:
- Meet demand on time and profitably
- If not possible, minimize the lateness
- Operate with minimum inventory
- Ensure high machine utilization
- Satisfy the minimum order quantities (MOQ) requirements
- Create long runs with minimum changeovers
- Follow specific rules like no changeovers on weekends or at night
- Honor the manufacturing budget set at the beginning of the period
These and similar objectives often conflict with each other. For example, nice long production runs to stabilize the machine also translates to inventory that one does not need at the moment. Similarly, MOQ means extra on-hand inventory. Further, the mathematical interpretation of some of these objectives is not always clear. For example, what is the difference between:
- A 1-unit order being 100 days late, OR a 100-unit order being 1 day late?
- Missing the target on 1 product by 100 units, OR missing the target on 50 products by 2 units each?
When we discuss this with businesspeople, one thing becomes clear: in the case of scheduling (and planning), they use the word optimal in a broader sense. While they will be happy to take a truly optimal schedule, they are doubtful that such a schedule exists or is implementable. Instead, they are focused on the ability to:
- Create a reasonably good starting schedule
- Integrate with data sources to eliminate manual work
- Find and mitigate trouble such as missing orders or too much inventory or a bad changeover
- Quickly see the connection between activities
- Ask and evaluate what-if questions
- Manipulate the schedule
- Drag and drop
- Increase/decrease quantities
- Add/delete new activities
- Leave room for heroic work factory workers sometimes do get the product out the door for the customer.
When seen holistically like this, optimizing the schedule takes on a meaning well beyond the mathematical meaning of optimal. As Ken puts it: “Optimal is the wrong term for scheduling… it is impossible to fully define the metric. It is also not art… a term an outsider uses. It does require harnessing community intelligence and a willingness of the team to work in the trenches as Karl Kempf has noted. Working knowledge requires living in the factory.”
Register now and join us for our next webinar on Wednesday, August 16th. I will be talking about what I have learned from the schedulers and planners that I have had the good luck to work with.