A project focused on automating decisions and improving efficiency.
The first part of the project focused on automating the plans generation by 1) replicating with algorithm the planners decision making process and 2) embedding validation steps into the process. The objective of this phase was to generate similar plans in a much reduced time. This first phase was successfully delivered, reducing planning time from 3-4 days to just a couple of minutes.
The second phase of the project aimed at leveraging optimization techniques to produce more efficient plans that reduce the overall transportation costs while keeping the level of service.
The planning engine focuses on three specific areas:
Orders Grouping: Which orders can go on each truck
Trucks Routing: Optimizing the delivery routes
3D packing: Matching the package size to the truck volume, specific to the exact placement, and considering multiple different truck sizes and limitations
The solution is delivering a 10% reduction in total transportation costs, sizably above initial expectations of 2% reduction. Additionally the system is providing a plan in a matter of minutes compared to the original 3-4 days. The next phase will be a global roll out.