We recently posted an article on applying six-sigma to your inventory and are now following up the series with another six-sigma related article. Today I will talk about utilizing six-sigma to assess the performance of all your suppliers in order to make sure your operations are not being adversely affected by them. You will also get a chance to read about how your supplier lead time performance can affect your inventory levels.
Before starting my discussion I wanted to point out that when I refer to the term ‘Six-Sigma’, I’m referring to the methodology of variation reduction in operations and any time I mention ‘Lean’, I’m referring to the methodology of waste elimination in processes.
Most if not all businesses have a decent network of suppliers that provide them with parts, be it finished goods or raw materials, to run their operations. Often times problems in operations and manufacturing are traced back to the suppliers, but it is typically the result of a fire-fighting approach instead of a pro-active approach. Many businesses end up blacklisting some of their suppliers for failing to meet the proper quality standards. In this blog post I will identify how businesses can detect supplier problems earlier through the use of six-sigma methodologies and instead of black-listing their supplier, can help them improve their quality standards.
Six-Sigma and Supplier Lead Times
Lead time is a big metric that supplier performance is measured on. Many organizations tend to average up the lead time for their suppliers at the end of the year and kick out the suppliers that are not meeting the lead time metric specified by the organization. Of course, lead time is not the only metric taken into account but it is definitely a very important metric for judging your suppliers. Let’s take an example of two suppliers, call them Supplier A and Supplier B that provide widgets to organization X.
Organization X has a lead-time metric of 30 days for each supplier, meaning that at the end of the year the average lead time of the suppliers needs to be a maximum of 30 days for parts delivered. Now say Supplier A delivers the widgets at an average of 38 days with a standard deviation of 6 days. This supplier delivers parts within 32 and 44 days 67% of the time, and between 20 and 56 days 99.9% of the time. The supplier does not meet the organization’s requirement but has a ‘statistically controlled’ lead time distribution.
Supplier B on the other hand has an average lead time of 28 days however, the supplier’s lead times are distributed all over the place. Sometimes the supplier delivers parts in 14 days and other times the parts are delivered in 56 days. The supplier’s lead times are not in ‘statistical control’ and therefore unpredictable.
Which of the two suppliers would you go with?
Supplier B seems to meet Organization X’s lead time metric for average lead time but Supplier A is more predictable even though it fails to meet the average lead time metric.
I would argue that out of the two suppliers, Supplier A would be more beneficial for Organization X compared to Supplier B assuming all other factors stay the same. Using principles of six-sigma, you can see that Supplier A has its processes under ‘control’ and the lead time follows a ‘normal distribution’. Supplier B on the other hand doesn’t seem to have its processes under ‘statistical control’ and the lead times don’t follow a predictable path.
Lead Time & Inventory
This is where I will briefly introduce the impact of your suppliers on inventory. Say Organization X works with Supplier B that has a highly volatile lead time. The inventory would follow similar fluctuations as the lead time of the parts. Sometimes organization X would have too many widgets in stock and other times they would be facing stock out. In order to compensate for the lead time volatility, organization X would choose to overstock on inventory and at the end of the year this would cause a bunch of parts to be scrapped to clean up the accounting books.
However, for Supplier A, if organization X used a six-sigma approach it would know that 95% of the time the lead time of widgets will be between 26 days and 50 days. They can then stock adequately based on the worst case scenario of 50 days and come the end of the year they would not have to scrap as much as they would for Supplier B. What would be even better is if Supplier A could reduce their lead time variation by improving their internal processes.
Organization X should work closely with Supplier A to reduce the variation in lead times and jointly figure out the root cause of why the average lead time is high. Lean Six Sigma teams might find that the root cause is as simple as the time taken to get the paperwork done. If the process is in control there needs to be a systematic change in the way things are done at Supplier A to bring the lead-time down and to reduce variation. One of the questions organization X needs to ask themselves is if the lead time metric is even feasible. Too often organizations focus more on averages to determine compliance instead of taking a holistic approach to understand the overall performance of their suppliers. Applying six-sigma can give organizations a much better way of assessing supplier performance.
Case for Supplier B
I do have to say that Supplier A might not always be the best solution but it is a more predictable option. Supplier A has a controlled system which means that in order to improve the lead time variation and average, they would need to change or upgrade the system. Sometimes this could be as hard as changing management’s mindset which can be a pretty challenging task.
Supplier B on the other hand seems to have more ‘special cause’ circumstances affecting its performance. Oftentimes these are low hanging fruit and can be fixed quite easily. In that case supplier B’s lead times could be brought in ‘statistical control’ by fixing the special cause circumstances and could even perform better than Supplier A once in ‘statistical control’. Hence, even though I would strongly recommend Supplier A as the one to focus on, there are ways to improve Supplier B’s performance and get them up to speed.
In conclusion I would like to state that from a pure lead time perspective, it is more beneficial for organizations to pick suppliers with predictable lead-times even if their average lead time is high instead of suppliers with unpredictable lead times even if the average falls under the defined specification. The more predictable a process, the easier it is to improve and even automate it. It should be the goal of organizations to establish a level of predictability to all their processes in order to truly make improvements.
The goal of RigBasket is to make the application of six sigma as easy as possible so our clients don’t have to worry about the complex calculations and statistics that I mentioned above. Schedule a demo with us to see how we can help improve your supplier performance.