Who wouldn’t prefer to wear a custom-tailored suit over something bought off the rack? Especially if it can be had for the same price, or even cheaper? In much the same way, we find that companies have a taste for supply chain analytics that are carefully tailored to their own business, quirks and all. In this talk we will discuss supply chain analytics broadly, provide some examples, and then address conditions when a custom approach to creating a supply chain decision support tool makes good sense.
This is a reference flow chart – the idea is a collaboration so that demand and supply can, indeed, be matched. Most businesses have settled on calling it some close variation of S&OP – Sales and Operations Planning. What can we do to shape demand? Slow down sales to push it out? Offer a discount to pull it in? How about the supply side – cancel orders? Expedite?
We focus a lot on planning, which is what you do with a forecast. Forecasts change all the time; your plan doesn’t have to. One of the keys to planning is to have something of a playbook to follow. What are we going to do if orders don’t materialize? What if that big government order comes in?
There are some really neat analytical techniques that can be deployed in each of these boxes. However, one of the hardest challenges – and one with great payback – is simply consolidating the data on past demand and anticipated future demand. Simply being able to see what it looks like gives rise to good discussions. This is a case where underlying analytics enable a human discussion and decision process.
This process has been followed in the tech world for years, and we’ve worked with several clients who are essentially tech players in the healthcare space. All have benefited form this ability to blend views from the demand side (e.g., regional demand, new launches) and the supply side (capacity, constrained supply, etc.).
The techniques translate easily to the pharmaceutical manufacturers and distributors.
This picture emphasizes the importance of data visibility without particularly fancy analytics.
This is what the S&OP discussion is all about. Different parties have different perspectives; why do they differ? What do you know that I don’t know? In this example, why is it that the sales force is eternally optimistic, while marketing thinks demand is going down? I should add – this is real data!
Folding in a statistical “reality check” helps to ground people, and taking pains to identify the uncertainty is especially important for a quality conversation.
Sometimes you just ID bad data … typos, crazy forecasts … and correct them. Other times there are major disconnects between organizations that need to be resolved.
Again, note the use of spreadsheet tools to make data accessible and visible. There’s some analytical rigor behind the scenes, but much of the value is in just getting the story the numbers tell in front of people. Other tools will help draw attention to the elements seemingly requiring the most attention – a sorted action list, really.
Note the ability to click to roll up and roll down. This allows you to look at your data from different perspectives – perhaps showing how demand patterns are evolving at each of several regional hospitals, then turning it around and looking at it from a supplier-centric view.