Forecasting demand in aerospace is difficult because demand is both highly erratic and intermittent. Also, TAT (turnaround times) need to be forecasted as well. In practice, probabilistic forecasting is required to achieve any meaningful results in the aerospace industry.
Nowadays it is often taken very much for granted that we will be able to go to the airport, then sit on a plane to arrive safely and on time at our destination. However, behind the scenes there are so many factors that ensure this not only happens safely for the client, but also profitably for the companies involved. In this episode of LokadTV, we discuss the aerospace sector and the key features of this market that make it so unique from other businesses.
Nowadays, aircraft are made of up to 300,000 pieces and airlines need to have them fly as much as possible to be profitable. As a matter of fact, it costs up to $200,000 per day to ground a plane for maintenance. This makes on-time aircraft movement a critical success factor that drives both costs and revenue. We understand how airlines manage their stock and even collaborate with each other to ensure that losses due to aircraft on ground (AOG) are kept to a minimum.
In addition, we explore how airline maintenance works and the budget constraints that affect even the largest aerospace companies, who can own over a billion dollars of inventory each. We also discover how airlines manage to work together, even in a highly competitive world, in order to keep aircraft flying.
Finally we wrap things up by explaining why classical supply chain models can’t be applied to aerospace supply chains and why a probabilistic approach that completely embraces the fact that your supply chain is built on loops and not on a direct flow from producer to consumer, is actually what works best. We also go further into seeing what a supply chain solution for aerospace should try to achieve in terms of KPIs.
00:33 What are the key features that make this market so different?
02:02 How does the maintenance work?
03:02 How many components are we talking about here? What is the scale of operations?
03:54 What are the complexities that we associate with this supply chain?
05:59 How many parts on a plane are actually repairable and how many are just fully replaced with new parts?
07:43 The idea behind a plane is that it is going to spend more time in the air than on the tarmac. Does this mean that airlines have to hold huge amounts of stock?
09:48 What are the options that are open to an airline company if one component is missing?
11:30 Do the airlines work together? Surely it’s in their interest to see a competitor fail and lose revenue?
12:20 With hundreds of thousands of components, repairable loops, collaboration between different companies, how can we actually even begin to approach forecasting for this sort of scenario?
17:08 We have so much erraticity with this repairable process. How can you approach that? How can you know if something is going to be repaired in 2 weeks or in 2 months?
19:34 Why airlines can’t just have lots of parts in stock?
21:20 What are some of the main challenges that you have faced with forecasting for aerospace clients?
26:27 From an optimization perspective, what are you trying to optimize? Dollars, time?
28:24 To round things up, what is the key lesson you want to take away from today’s discussion?