Systems integration used to be the dark side of IT. Unglamorous. Underestimated. A greedy consumer of limited project resources.
Much has changed since then. Thanks to the progress made with middle-ware and standards leveraging XML, enterprise application integration, and web services, systems integration now takes a fraction of the time and effort that used to be required.
So why do some people still believe that integrated all-in-one systems are somehow better than best-of-breed ones?
I wish I knew.
What I do know is that tight integration isn’t all it’s made out to be.
Before I explain why an integrated solution that couples ERP and Advanced Planning and Scheduling (APS) too tightly can turn into a nightmare, let me say – right away – that integration has its advantages. It improves productivity by eliminating the need for tedious data copying. And it improves accuracy, as data is only entered once.
So yes, of course your APS solution should be well integrated with your ERP system, and this can be accomplished quite easily nowadays.
Tight integration, on the other hand, is quite another matter.
I’ve been involved in many ERP-APS integrations and most of the time we had to add options to overwrite data coming from the ERP system. This probably sounds bizarre: Isn’t an ERP system supposed to provide the most up-to-date, accurate data?
Well yes it is, but life ain’t like the movies. There are often limitations, assumptions and latencies that disconnect the system from reality, and then which do you believe?
In most production environments, reality wins hands down. I’ve actually seen planners start their day by overriding stock data with a spreadsheet from the warehouse because the system ‘didn’t know’ about expired stocks or palettes just out of quarantine.
There’s more.
In a tightly integrated system, data is shared by everyone in the company. That’s the beauty of it – and the curse. In one unforgettable case, the planners needed monthly updates of the machine production speed to plan accurately. The finance department wanted no such thing, claiming that the information was unnecessary and likely to introduce unwelcome variations in costs. A quarterly average value was enough for them to track costs against budget.
Perhaps the most unfortunate result of tight integration is the way it limits your ability to explore what-ifs. The precious time saved by an APS system should be used by planners to explore alternatives. After all, isn’t moving from reaction to anticipation the essence of planning?
Creating what-if scenarios involves ‘playing’ with data to assess the impact of various changes. For example, you might need to tweak your master data (What if I introduce a new product?) or transactional data (What if demand increases next week?), but such changes cannot be made in a tightly integrated system where the data has to reflect your operational reality.
So here’s the crux of the matter. If you’re thinking of making the big leap from spreadsheets to an APS system, make sure you aim for the best of both worlds: the power of integration and the flexibility of a planning bubble where you can explore alternatives without worrying about how they will affect the rest of the system.