Forecast

    I would like to start by saying that planning lead time is not a popular parameter, very few people understand the impact of ignoring the lead time when it comes to computer parameters and as a planning tool.

More than ever, the pandemic, highlighted the impact of ignoring the parameter. In the MRP world there are two kinds of master production schedulers, the ones who love "safety stocks" and the ones who understand "lead times". The first category are those who suffered from shortages in the past and at night, still hear the screaming and yelling from the people needing the material. On the other hand, very few people rely on lead times to ensure a healthy inventory level.

In order to understand the difference, we need to clarify what happens when the safety stock and the inventory level is not the same or more. Internally the system will send signals to the planner or planners to make sure those items are addressed. This will create a list of items in desperate need, but the list can grow exponentially making the people numb to those signals. Usually the root cause for low inventory levels is not understood and in some cases the reasons are not worthy of investigation due to more important shortages. At this point a "hot list of parts" is essential for the business to operate. Supplier performance, quality rejections, transportations delays, rework process to name a few, are contributing factors to the insufficient safety stocks. Real data from ordering to arrival it's a must.

A few years back, "plan for every part" was a very popular topic, all related to inventory level controls to ensure on-hand material.  It's a simple concept but not easy to tackle, when your number of different materials or components is in the thousands, you need a small army of people. While is not impossible, the main reason to make this a top priority is to realistically state the lead time from the supplier to the stocking process. Lead times can be manipulated, in most cases, you want the lead time on the system to be larger than the real one, this way, not only the material is available, but the supply chain pipeline is healthy.

We were about to talk about Forecast, why the deviation on the topic? Very simple, in the real world you want to overstate your forecast versus actual/future sales because of the material lead times. Now the question is how much more?, the answer is also quite simple... look at the sales history!. Unless is a brand new product and nobody knows how the market is going to react, 9 out of 10 times, the history will give you a very good idea on what to use as forecast. Yes, there is variation due to seasonality or variation due to competitors, but when the product has been on the market for quite some time, you can tell by looking at the data where the number is going. Just because the final forecast and ordering process resides on the manufacturing areas, make no mistake, this is not a manufacturing exercise.

Inventory policies, projects to reduce inventory, and cash are just a few important factors to consider when you forecast. But chasing out-of-stock inventory because of real customer orders, it's a very dangerous situation, it leads to horrible habits creating all sorts of problems and behaviors not easy to revert. Expedites for the most part, reduces the profit margin instantly.

Nobody has a crystal ball predicting the future, forecasting is not an easy task, but neither is manufacturing!