Automated Transportation Routing Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle routing and pickup/delivery optimization systems face challenges in efficiently decomposing large problems into smaller, tractable instances, often resulting in partitions that are difficult for users to understand, leading to a lack of trust in proposed solutions.
Innovation Solution
An automated method for partitioning transportation routing problems involves determining a threshold number of shipments per partition, selecting a geographic center, mapping sites, and radially aggregating them to create partitions with a maximum of the threshold number of shipments, using heuristics that match real-world constraints and user mental models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated decomposition strategies are used to efficiently solve large transportation routing problems, then the problem-solving efficiency is improved, but the partitions become difficult for users to understand and trust
Solution Approach 1:
The patent applies segmentation by dividing the large transportation routing problem into multiple smaller sub-problems or partitions. Each partition contains a manageable subset of shipments and vehicles, allowing efficient solving while maintaining overall problem coverage. The segmentation is achieved through clustering algorithms that group geographically proximal or temporally related shipments together.
Solution Approach 2:
The patent introduces an intermediary layer between the automated decomposition process and the user. This intermediary provides explanation mechanisms that translate complex partitioning decisions into understandable rationale, such as showing geographic boundaries, temporal windows, or constraint-based reasoning that justifies why certain shipments are grouped together.
2Ease of operation
If manual decomposition is used to create user-understandable partitions, then user understanding and trust are improved, but the problem-solving efficiency deteriorates
Solution Approach 1:
The system applies self-service by automatically generating partitions without requiring manual user intervention. The decomposition process autonomously analyzes the problem instance, applies clustering algorithms, and produces partitions that are both computationally efficient and explainable, eliminating the need for users to manually define partition boundaries.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system provides users with explanations of the automated partitioning decisions. This feedback includes information about the criteria used for partitioning (geographic proximity, time windows, vehicle constraints), allowing users to understand and trust the automated process without manual involvement.
3Productivity
If existing automated partitioning methods are used, then computational tractability is improved, but the partitions do not match user mental models
Solution Approach 1:
The patent changes the parameters of existing automated partitioning methods by incorporating multiple dimensions for partitioning: geographic location, time windows, vehicle capacity constraints, and shipment characteristics. This multi-parameter approach creates partitions that align with how users naturally think about logistics problems while maintaining computational efficiency.
Solution Approach 2:
The patent applies dynamics by making the partitioning strategy adaptable to different user needs and problem characteristics. The system can dynamically adjust partitioning criteria based on the specific problem instance, user preferences, and operational constraints, allowing the same automated method to produce mentally-aligned partitions across diverse scenarios.
Data Source
AI summary
Method and system are provided for automated partitioning of transportation routing problems. The method includes: determining a threshold number of shipments per partition; selecting a geographic center; mapping delivery and/or pickup sites at geographic locations; scanning radially around the geographic center to determine the sparsest or densest region of sites and selecting a starting point in this region; and progressing from the starting point radially around the geographic center aggregating sites into partitions with a maximum of the threshold number of shipments in a partition. The method may include: solving each partitioned instance of a problem to generate one or more optimized routes; and creating a union of all the instances solutions.