As someone who spends a lot of time talking with manufacturers that are trying to improve and streamline their operations, one of the questions I hear quite frequently is, “How do I define application ‘footprints’ between corporate planning and the shop floor execution?” Specifically, the confusion lies in deciding which processes should reside in an Enterprise Resource Planning (ERP) system versus a Manufacturing Execution System (MES). Also, this evaluation typically includes discussion on where to store the “master” data.
Unfortunately, there is no panacea that fits all cases. The answer depends on how much process flexibility is needed, how much complexity exists in your manufacturing processes as well as what IT systems already exist. We all operate within a set of budgetary and “change” constraints. The challenge is how to best plot a path that can lead you to a better solution tomorrow than what you currently have today.
As a way to help you in your journey, I would propose that you operate somewhere along the two ends of the spectrum outlined below:
- Highly Static IT Architecture – this type of organization strives to “lock down” IT systems and processes across their enterprise; the driving goal is control and rigidity in systems architecture. This type of organization has most likely already implemented ERP across corporate functions, but is now struggling with how to best extend this control down to the plant floor.
- Free-for-all IT Architecture – this type of organization lets plant managers self-manage their IT applications. ERP likely exists, but most probably hasn’t been implemented down to the shop floor. The challenge here is that after 15-20 years with this type of structure, waste is inevitable. Duplication of processes has created more work, less accuracy and manual “work around” processes. Supporting and maintaining these IT systems has now become a huge problem.
The first point to consider is that plants are heterogeneous, so have different systems and equipment. Integrating a complex ERP application to 20+ different sites with different configurations will ruin your budget. Based on what I have observed, a best practice is to avoid making changes in ERP once it has been implemented, as change is typically difficult and expensive. Processes that require frequent change are best served by an MES. Master data can then be located in the system that most frequently accesses that data.
If you have several locations, then maybe now is the time to standardize your MES to become an enterprise solution, one that can be centrally managed (a Center of Excellence works great) yet still supports local idiosyncrasies. This type of solution now exists. Then, you can establish a consistent “plug in” at every plant that your ERP can easily connect to. This strategy allows your ERP to remain static, eliminating hassles, challenges and costs in achieving your global roll out vision. Meanwhile, your MES can perform the functions it was originally designed for – execute the processes that were planned in ERP – while still supporting process changes as often as necessary.
Hopefully, this explanation makes sense! If not, feel free to drop me a note or send me a comment to this post, and I will be happy to explain further.