Dynamic Zone Configuration for Supplier Assignment and Travel Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for assigning suppliers to destinations result in excessive travel time or distance due to sub-optimal zone configurations that are static and not updated as destinations are added or removed, and fail to consider supplier locations when defining zones.
Innovation Solution
A system that generates dynamic zone configurations using machine learning to group demand nodes and assigns suppliers based on travel times or distances, considering both demand and supply node locations, and selects the optimal configuration by minimizing total travel time or distance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional algorithms for solving Transportation Problem and Traveling Salesman Problem are used, then optimal assignment and routing can be achieved, but computational cost becomes excessively high when the number of origins and destinations becomes large
Solution Approach 1:
The patent divides the set of demand nodes into multiple zones using partition clustering algorithms (such as k-means or hierarchical clustering). This segmentation reduces the complexity of the assignment problem by grouping demand nodes into manageable zones, which then can be assigned to supply nodes more efficiently. The segmentation transforms the original complex optimization problem into a series of simpler sub-problems.
Solution Approach 2:
The patent implements dynamic zone reconfiguration where zones are periodically reassigned based on changing demand patterns, supplier locations, or other parameters. The system dynamically adjusts zone boundaries and assignments using optimization algorithms to adapt to evolving conditions, thereby maintaining optimality without requiring complete reoptimization of the entire system.
2Ease of operation
If static zones are created to simplify supplier assignment, then assignment process becomes easier, but zones are not updated as destinations are added or removed resulting in excessive travel time
Solution Approach 1:
The patent transitions from static to dynamic zone configurations. Zones are periodically reassigned based on current demand node locations, supplier positions, and other relevant parameters. The system uses optimization algorithms to dynamically adjust zone boundaries and assignments, ensuring that zones remain optimal despite changes in the operational environment.
Solution Approach 2:
The system incorporates feedback mechanisms where actual travel data, demand patterns, and supplier performance metrics are continuously monitored and fed back into the zone assignment algorithm. This feedback loop enables the system to learn from actual operations and refine zone configurations to minimize travel time and improve assignment efficiency.
3Loss of time
If conventional techniques allocate suppliers to pre-defined zones considering supplier locations, then some travel optimization is achieved, but zones are defined without considering supplier locations resulting in suppliers traveling to zones further from their starting points
Solution Approach 1:
The patent implements a dynamic zone definition process where zone boundaries and assignments are continuously adjusted based on real-time or near-real-time data about supplier locations, demand node positions, and travel conditions. The system dynamically reoptimizes zone configurations to ensure that suppliers are always assigned to the most efficient zones given current conditions.
Solution Approach 2:
The system changes key parameters such as zone boundaries, zone centroids, and assignment criteria based on varying conditions including supplier locations, demand patterns, and travel costs. By making these parameters dynamic rather than fixed, the system adapts zone definitions to minimize travel distance while considering the actual positions of suppliers and demand nodes.
Data Source
AI summary
A computer-implemented method includes generating a plurality of zone configurations, at least by applying a machine learning (ML) model to demand node data indicating locations of a plurality of demand nodes; determining a plurality of respective costs for the plurality of zone configurations; selecting a particular zone configuration, from among the plurality of zone configurations, based on the plurality of respective costs; and storing in a memory one or more data objects representing the particular zone configuration.