This is not specific to semiconductor, but I think the principles will apply universally. Much of this is feedback/lessons learned from an existing customer (who gave consent to share their experiences) using RapidResponse for Master Scheduling.
- If you are not already running your business on strong MRP principles, you will likely spend more time than you expect getting your data trustworthy enough to do your Master Schedule via RapidResponse. Then you will spend a ton more time getting your business to think like MRP so they understand how to do their jobs and take advantage of the great features RapidResponse offers.
- RapidResponse is only as correct as its input and _multiple_ (business) functions need to agree on the information that creates your MPS (S&OP anyone?). Master Production Scheduling can be a challenge because forecasters want to forecast one time for the season yet still expect Supply Chain to have all the right product planned/produced as they want it throughout the season. If you are going to use RapidResponse for a Master Schedule, start with an educated forecast. Use a statistical model or the best-educated guessers in the house! This is not about holding anyone’s feet to the fire for an accurate forecast. It is about using your best guessers when you already know a forecast is a guess and need the most meaningful forecast.
- Do not disconnect your aggregate from your detail. It can be tempting to get management’s numbers and then do loose math to see if it looks like your MPS is on track, but then why bother using RapidResponse? One of the beauties of RapidResponse is its ability to think about many related data points, and show you the impact. Take advantage of it by connecting the levels of your data so you know you have a Master Schedule that the sourcing and production teams can support, while still meeting aggregate targets. Even if you need to make educated guesses in your data to connect your aggregate and detail data, make sure they ARE connected. You cannot expect to compare apples to oranges, yet have advanced warning that there are not enough apples.
- Master RapidResponse’s hierarchies – when doing master scheduling, you often need to move up and down levels, as well as across levels to check the impact in aggregate, in detail, and then across multiple categories (how do totals for this category look for ABC? How about XYZ?). Smart hierarchies and user mastery of their use is a very low-cost, easy to implement way to get a lot of bang for the buck out of a workbook, though it can take time to get buy-in for someone who is used to using Excel auto-filter.
- RapidResponse is a big number-crunching system so it may be tempting to shorten your frozen zone once you have RapidResponse in place – don’t! Unless you are starting with very poor liquid, slushy, frozen time fences, enforce your existing fences for a period of time until you have “normalized” your business to RapidResponse’s MPS capability. Only then will you be able to accurately assess the impact of changing your frozen zone.
- Understand MRP and define your ideal MPS process via MRP (specific to your business NOT a tool). Then determine where RapidResponse will do a good job supporting your processes and where your processes might be able to adjust to take advantage of RapidResponse’s strengths/capabilities. Two things here:
- Maximize what RapidResponse is good at rather than forcing it to behave in a way it does not do well. If you focus on forcing RapidResponse to fit your way of thinking (i.e. high levels of customization) you are committing yourself to a higher maintenance cost AND limiting your ability to take advantage of other areas of RapidResponse later.
- Identify your “big rocks” so you know what is most important to your process at the start and what you can give up when you need to compromise. What are the “hard” parts about managing an MPS that you hope RapidResponse will resolve? Remember, you don’t need perfect – you need substantially better. Don’t try to solve/implement everything at once, but instead focus on low-cost wins and your important “big rocks” and make sure those needs are met.
I hope this helps!