On time delivery (OTD) and on time in full (OTIF) metrics are universally used. They also can be misleading measurements of supplier performance. Why haven’t these metrics evolved?
          When I was a kid, I was a big collector of baseball cards. I loved looking at the statistics on the back like slugging percentage and figuring out how it was calculated. Since then, a lot of advanced metrics have been created to try and sum up how players perform. OPS (On Base+ Slugging) made it onto baseball cards in 2004 and WAR (Wins above Replacement) in 2014.
          A lot of critical thinking and data crunching was done around baseball to try and get more precise statistics. Baseball is a huge business, and it is critical to win, which is the same for manufacturing. Yet, in manufacturing when we measure the performance of our supplier base, we are still using the old, tired, and imperfect OTD/OTIF. Why?
          Measuring OTD/OTIF is answering a yes or no question when we have the data that can tell us how much instead.
          Here’s an example. Assume we have data on the last 60 shipments for five suppliers. Below is a chart of the OTD%.
          Based on this data, it’s clear Supplier 2 is our best supplier at 90% OTD and Supplier 4 is the worst supplier at 50% OTD. Let’s give Supplier 2 our gold supplier award, and let’s look for a secondary supplier for Supplier 4….
Except, the results of this attribute data are not painting a complete picture. Here are 5 histograms to show how late they were.
          Looking at the histograms, you can see that Supplier 2 might not be your “best” supplier, because there is more variation in their delivery dates. In the table below I added the metrics, average days late and the standard deviation. These are some alternative measures to just OTD/OTIF%
          These metrics are better expressions of the true supplier performance and how much of a disruption it is when they are late. This sense of magnitude can help us make better decisions. Looking at the new metrics, you can see even though Supplier 4 has a 50% on time delivery they are the most consistent supplier with the least variation.
In production planning variation is your biggest enemy.
          Are these metrics all we need to help our production planning/scheduling team? Maybe not. If we are trying to hit our customer’s due date, we need to have some level of certainty of how late material could arrive. That way our production planners could use a safety lead time to keep production on time. Do we need 95% confidence? Maybe 98% confidence? That is a unique business decision based on how critical the timing of the particular part is. Below is some data to show a potential maximum late number.
          These metrics might not be perfect for your business, but they are better than just relying on OTD%. Challenge your business to think critically and decide what metrics are right to drive better performance.
Message me if you would like to discuss KPIs further.