Case Study : Major Transportation Company Saves Big with Predictive
The client, a transportation and trucking conglomerate, acquired dozens of commercial-grade large truck brands. The business was booming, but a persistent issue arose that was hurting their bottom line.
Before we dive into the problem it helps to understand the business. The client basically operates like any major car brand. Dealers sell trucks and buses. Those vehicles are used by other companies to run their businesses and make a profit, and when those vehicles need service (like all cars need service for maintenance or repairs) the trucks are brought into a service center.
With so many different makes and models of trucks in operation, It was difficult for their service centers to keep needed parts in stock. There were more parts than there was space to store them. We're talking about thousands if not hundreds of thousands of parts. There is simply not enough storage space to keep every part on hand.
But here's the real problem, if a part is not in stock, the commercial truck remains out of service. If trucks are out of service, owners are losing money, an average of $4k/day, and their angst is directed at the truck manufacturer.
One of the great things about the trucking industry is the data. These trucks are not your mom's station wagon. They travel millions of miles, hauling thousands of pounds, and they're tracked the whole way. The client also kept meticulous service records for each truck. That means data points.
Using this trove of data, our team constructed life cycle graphs for each part used on each commercial truck brand. From there we developed a model to predict part failure. We applied the model to ALL client-branded commercial trucks currently operating. What does this mean?
The client is now able to predict the average failure time for each part and thus better stock part inventory to meet projected demand.
Needless to say, this solution exceeded expectations.
Previously, the client was forced to make little better than a guess regarding part inventory. Those results? Excessive wait times for service and angry customers losing revenue.
With STAND 8's predictive model, the client could anticipate demand and order parts before they were needed. This meant less time in the shop, more time on the road, and more profit for customers who could expand and buy, you guessed it, more trucks.