Service Supply Chain Expert: Measure Equipment Availability, Not Fill Rate

There’s a lot of rich history behind Servigistics, PTC’s service parts optimization solution. Vinod Arekar, a PTC Fellow and service supply chain expert, has been a huge part of the technology’s development, overseeing implementations in the U.S. Air Force, U.S. Navy, and LAM Research, among others.

I had the opportunity to speak with him about a LiveWorx 18 session he’s co-hosting with Dr. Andrea Lobo, CEO, SimAcumen, and Michael Mohesky, a supply chain specialist at Boeing. The session focuses on equipment availability as a guiding metric – why businesses should use it, and how to apply it in real life.

What inspired this session?

The reason you have a car is because you want a convenient way to get to work, go to restaurants – whatever. You’re not interested in whether a dealer will have an oil filter when you need one. Your concern is, how will changing the oil inconvenience me? How long will I not be able to drive my car?

Organizations stock hundreds of millions of dollars in service parts because products fail. Therefore, parts demand is associated with failure. Customers don’t like failure because that means they can’t fly passengers, conduct medical procedures, produce semiconductors, and so on.

Equipment availability addresses this problem because it’s designed to minimize the amount of downtime customers experience. Fill rate isn’t, but it’s one of the most commonly used metrics among service parts planners. Fill rate ensures you have parts available after something breaks. It doesn’t give you credit for making the equipment more reliable so it will have fewer failures, whereas equipment availability does.

Many industries haven’t implemented equipment availability, though they have realized the need. Defense OEMs and agencies are the exception. Several of our clients in this space use our equipment availability algorithms – Lockheed Martin, the U.S. Navy, and the U.S. Air Force, to name a few.

Why the hesitation to move toward equipment availability?

Well, people assume that it’s going to take a lot of data and complexity to implement, but in practice, that’s not the case. Equipment availability isn’t that difficult to model, and it only needs a few data points:

  • Your installed base – equipment BOM and location.
  • How customers are utilizing your equipment.
  • Which parts are essential for the equipment, and which are required in the process of repairing the essential parts, if any.

Given that we live in the world of IoT, accessing this information isn’t an obstacle anymore. Most organizations simply need to take the next step and apply that data to their parts planning. I actually touched on this in an article I wrote not long ago:

spare parts planner
Arekar discusses how using IoT data and predictive analytics can reduce service parts stock.

How does a focus on availability change the day-to-day jobs of service parts planners?

It changes the parts mix they must stock. In the past, they may have organized inventory to achieve a 90% fill rate, but if their key metric is equipment availability, the parts they need to stock and the quantities will be different, and some of the changes may not be intuitive to the planners.

Obviously, this can be hard to adjust to. Changing the mindset involves building models of availability-driven service supply chains and then running simulations to validate that the equipment uptime will achieve the target.

Most of our customers in this space run simulations before implementing the solution. In some cases, your business may have additional complexities such as redundancy, or may be working with a different set of assumptions than what the optimization algorithm models. In such cases, use simulation to tweak the solution and make it applicable to real-world situations.

What advice would you have for companies that want to target equipment availability, but are currently working with part fill rates?

It depends on what their operations look like. There’s not a cookie-cutter solution. Generally, I’d say make sure you have access to your IoT data. Again: equipment location, number of assets in the field, and utilization. If one part of your business is more data-ready than others, do a pilot implementation with that business area while you collect data for the rest of the business.

Second, develop availability models, and then run simulations. Once the simulations show that whatever availability algorithms you’re using are consistent, you can implement them in your operating environment.  If you are skeptical and feel that you are operating to a good fill rate (so what’s the need for availability?) just run your current stock levels through a simulation and do an objective evaluation – you will be happy that you did!

Arekar is co-presenting on this very topic with Dr. Andrea Lobo, CEO, SimAcumen, and Michael Mohesky, a Supply Chain Specialist with Boeing. To learn more about their session and others relevant to aftermarket service, explore the event's service transformation page below: