Why pragmatism matters when scheduling high-mix low-volume production

I've been implementing scheduling software for manufacturing companies around the world for many years now. As a German statesman, I am expected to be particularly accurate, especially by international customers. To the surprise of many people, I proclaim the opposite when it comes to scheduling. Therefore, this blog explains why pragmatism is - in my view - a good advisor in terms of scheduling a high-mix low-volume business.

It starts with the evaluation and selection of the scheduling software. I often experience that companies, that are looking for a replacement of their xls-based disaster, load the desired requirements into a long list of features. Potential tools are then evaluated using the related checklist. Candidates who do not meet all criteria are eliminated.

This then leads to an endless search and being left with the unsuitable xls solution in the long run (please read also The vicious circle of getting SMB production scheduling started). Therefore, in my view, pragmatism is necessary here to move forward:

• It makes no sense to create a rigid list of requirements based on experience with an unsuitable tool. Of course, every use case requires certain “must-haves”, but you should also be aware that no – affordable - software will ever meet all current requirements. Therefore be open-minded and pragmatic with your list of requirements.
  
  
• The same should also apply to the processes related to scheduling. Experience has shown that the attempt to “pour” the current processes one-to-one into the new/desired tool fails. And here, too, it makes sense to be pragmatic and to follow the structures and processes of the scheduling approach rather than the other way around.  

In summary, it can be said that those who do not lose themselves in the details of a rigid and long list of requirements are successful, but rather those who pragmatically follow the approach of an appropriate solution.

But pragmatism is also an issue after selecting when using scheduling software:

• For example, in the implementation process, a decision has to be made as to how detailed job master data should be predefined.
  In general, the possibilities range between a very detailed, deeply nested approach that provides predefined job routings for all eventualities and focusing on the usual 80% of customer orders at a sufficiently precise level. Here, too, I experience again and again that customers have very extensive and long discussions, only to conclude that more details promise more scheduling success. Here I always vehemently disagree and recommend a pragmatic approach with little depth of detail, but which allows a quick go-live.
  
  
• Defining resource master data within the launching process is another example.
  Here I experience that a lot of time is used to think about resource productivity. While the finest gradations are often desired, I usually advise not to make a distinction at all unless there are productivity differences that have a significant impact on total job throughput times.
  
  
• But the most important example in this context is when it comes to quantifying runtimes.
  Here customers have the utmost respect – if not to say fear - to estimate runtimes. This inevitably leads to the striving to determine runtimes as elaborately and precisely as possible. From my point of view, this is mostly completely unnecessary. My experience shows that the first intuition of a production planner is in 95% of the cases more than sufficient for an initial quantification. And that's exactly what I advise: be pragmatic and think straight  

All these examples have one thing in common: they are about data accuracy. This is a very important aspect, especially for job shop manufacturers. Here the environment is very dynamic, and the data is correspondingly very volatile. Consequently, a planning board in such an environment will never (!) show the 100% truth. Accordingly, it makes no sense to strive for 100% perfection. Rather, the pragmatic but realistic goal must be to gain transparency about the shop floor activities and to obtain reliable delivery dates with sufficient accuracy. And this cannot be done on a minute or hourly basis but rather on a daily or weekly basis. To wrap this up: I can state that I am (as a German statesman also gladly) accurate in many areas, e.g. in financial accounting. But my experience in the area of scheduling and here in particular in the environment of high-mix low-volume manufacturers shows that pragmatism is not only a good but, strictly speaking, the only successful advisor. Or in a nutshell: "Don't hesitate, do it".

Further readings

10 Machine shop scheduling best practices 

What is job shop scheduling