Skip to Content
Author's profile photo Former Member

building SAP supply chain KPI’s that work and support decision making

In my opinion, good KPIs help you make good decisions. And the decisions you make, can be the setting of a replenishment or planning policy. I’ve talked about policies in previous blogs and am of the opinion that these are at the center of an effective and automated SAP supply chain. Policies determine the order frequency and lot size of replenishments and, if properly assigned, guarantee a great service level whilst avoiding stockouts.

Therefore it is imperative to have a solid set of KPIs available for your decision making. Here are three very important ones:

DpA – dead stock value as a percentage of average stock. Dead stock is that portion of the inventory that does not get used over a defined period of time. By itself it doesn’t mean much. However, if you relate it to your average inventory value, it becomes a good indicator to show that when 75% of your average inventory remains in dead stock, you’re not consuming as much as you are bringing in. Dead stock as a percentage of average inventory should not be more than 25-30% and is ideal at around 15%. So when you find yourself above that threshhold, it is time to look for a different replenishment policy – one that works better for that item and helps bringing the inventory down.

FR – fill rate. The fill rate indicates the percentage of sellable product that was delivered to the customer immediately from available inventory. As per that definition, one should not measure the fill rate for MTO or ATO or FTO products since those should never be in free available inventory that can be shipped to the customer (they are made or finished to ORDER). Therefore it is important to segregate the MTS products from the rest and measure only MTS with the KPI “fill rate”. Of course the availability check should be set up accordingly. An MTS availability checking rule, in its purest sense, should simply check for available inventory (including safety stock) and if it does not find any, it should not confirm the sales order. Of course in some businesses you may want to look if there is anything in transit or on the production lines and confirm to the expected receipt date. However, for purposes of the fill rate measure, these delayed deliveries degrade the KPI. Only deliveries made immediately and directly out of available inventory account for a good fill rate. This is how you report on how good your planning and forecasting was for Make To Stock products.

FR – Flexibility Rating. Many people call this measure ‘Service Level’. I want to call it differently because it rates how flexible a company services their customer’s wishes. Make To Order, Assemble To Order or Finish To Order processes strive to react flexibly to fulfill a customer’s request. In that case you do not plan an inventory of a certain, specific product ahead of time (with a forecast), but you wait until the customer tells you exactly what they want – and only then you start your production or assembly lines. Therefore one should measure the time it takes to fulfill the request and how well that quoted time is adhered to. To get a good KPI here, you should set an estimated lead time (the total replenishment lead time in MRP3) and the availability checking rule should check with that lead time to quote a delivery date to the customer. The KPI should then  not only show how well we deliver to that quoted lead time, but also give extra points for early delivery.

As you look at these KPIs, you should be able to make conclusions about what policy might work to your advantage (to further automate but also improve on the KPI).

…and since it is a bit of a challenge to get these KPIs out of standard SAP, I want to point out a solution that I am endorsing very much, since I believe it will help you a great deal to construct your own set of KPIs that really work for you: It’s a product from Global Software called the Spreadsheet Server and works as an extension to Excel. It’s greatest feature is that no extract needs to be generated (as soon as you extract data out of SAP, the data is outdated), but the queries, which you define in Excel, pull the data directly out of the SAP table and leave it there.

…but I let Global Software explain how it works:

“Spreadsheet Server, Powered by the QueryExchange™ works as an intuitive Microsoft® add-in providing live reporting and analysis from SAP into Excel. This empowers end users with the ability to use familiar Excel features and functions to create dynamic dashboard views from SAP. These views can be accessed, automated and presented with our formula driven methodology, including summary balances and detailed drill-downs. The cloud-based QueryExchange™ component of Spreadsheet Server enables user organizations to select and tailor pre-built queries to meet their specific needs. Utilized in the QueryExchange™, the SmartPak™ feature is a way to group pre-configured queries together. Searching by selecting a SmartPak™ gives users a quick and reliable way to access a group of dedicated pre-defined queries. In addition, Spreadsheet Server has the ability to dynamically access the following modules: FI, CO, PCA, SD, MM, PP, QM, PM, HR, PS, New GL, and Special Purpose Ledgers. Spreadsheet Server enhances the SAP financial and operational user experience all while leveraging the significant investments made by the customer in Microsoft Excel.”

Assigned Tags

      4 Comments
      You must be Logged on to comment or reply to a post.
      Author's profile photo Kevin G
      Kevin G

      Regarding Dead Stock: How do you advise to determine the best time interval to use for the dead stock analysis?  The 3 month default seems too short.  In the aerospace industry we have some fairly long lead times, both internal manufacturing and external procurement.  Order lead times to customers could be anywhere from 2 months to 15+ months. 

      Author's profile photo Former Member
      Former Member
      Blog Post Author

      unfotunately, when you use SAP ERP standard report MC50 for a dead stock analysis, you are faced with very long wait times. Therefore it's not really feasable to run the report for many different time spans. Global Software (probably others too) have an interesting Add-On which extracts data from SAP (it actually doesn't extract but pulls data) and allows you to apply any formula you want. In that case you could pull all dead stock values into excel and select the time span there. An added advantage is that you could display the dead stock as a percentage of the average inventory, which makes it a much more valuable KPI.  http://www.glbsoft.com/sap/spreadsheetserver.aspx

      Author's profile photo Kevin G
      Kevin G

      Thank you for the response.  I understand the long wait times.  Running plant wide for 6 months is about 3min, 12 months runs for 7min.  Not too bad when I'm only running this once or twice a month.  I am aware of the available add-ons out there, but those aren't really an option at this time.  Even if we had the add-ons, we would still need to determine the appropriate time intervals to use.  I'm thinking that we may need to break out the runs for long lead time materials (maybe use 12 months) and shorter lead time materials (go with 6 or 3 months).  Thoughts?

      Author's profile photo Former Member
      Former Member
      Blog Post Author

      I do agree with you, Kevin. But deep down somewhere I feel that the dead stock value really doesn't have much to do with lead time (I know... this might be contrary to what I wrote before). Dead stock is a measure of whether you bring in too much or not. If you have long lead times, there is not really the need to bring in much, less frequently. You can have frequent orders that bring in little, often. When your consumption is regular and planable then you should have inventory levels close to zero, because you are dealing with little variability and therefore little chance of a big spike taking you out. That really doesn;t have much to do with lead time.

      I think the question really is: do I have enough movement (not a slow mover) to bring small quantities in frequently and for high moving items you don't need to look at a long period... If you loo at short periods to look at dead stock, you will not get good information on your slow moving items... but I'd argue that for a dead stock analysis you should look at high movers only... those are the ones with the biggest potential