Adaptive Route Reoptimization Using Precomputed Candidate Columns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current route planning systems for the 'middle mile' in delivery networks are computationally expensive and time-consuming, struggling to adapt to real-time changes such as ad-hoc additions and cancellations, leading to suboptimal resource allocation and efficiency.
Innovation Solution
The system utilizes a previously computed route plan and candidate column pool to reoptimize route plans in response to changes, reducing computational complexity and time by considering current tendering and execution status, allowing for finer granularity adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional route planning methods are used to generate optimal routes, then route optimization is achieved, but the computational time and cost become excessively high
Solution Approach 1:
The system pre-generates a comprehensive pool of candidate routes (columns) before the actual delivery execution. This preliminary action separates the computationally intensive route generation phase from the real-time delivery phase, allowing optimal routes to be prepared in advance while enabling quick adjustments when changes occur during execution.
Solution Approach 2:
The route planning problem is segmented into independent candidate routes (columns), each representing a feasible delivery sequence. Instead of solving one large optimization problem, the system generates multiple smaller, independent route options that can be evaluated and selected based on current conditions, reducing computational complexity.
2Adaptability or versatility
If traditional route planning systems attempt to handle real-time changes, then adaptability improves, but computational complexity increases significantly
Solution Approach 1:
By pre-generating a comprehensive pool of candidate routes that cover various possible scenarios and changes, the system is already prepared to handle real-time adjustments. When changes occur (additions, cancellations, modifications), the system can quickly evaluate the existing candidate pool and make adjustments without performing full re-optimization, thus maintaining adaptability while controlling computational complexity.
Solution Approach 2:
The system changes the approach from solving for optimal routes dynamically to selecting from pre-computed candidate routes. This parameter change transforms the problem from continuous optimization to discrete selection, significantly reducing computational complexity while maintaining the ability to adapt to real-time changes through selective re-evaluation of candidates.
3Productivity
If comprehensive route optimization is performed, then resource allocation efficiency improves, but the time to generate plans increases from hours to unacceptable levels
Solution Approach 1:
The system performs comprehensive route optimization in advance by generating a pool of candidate routes that incorporate various resource allocation scenarios. This preliminary comprehensive optimization ensures high resource allocation efficiency is built into the candidate routes, while the actual plan generation time is reduced to selection and evaluation of pre-optimized candidates.
Solution Approach 2:
Instead of generating one unique optimal route, the system creates multiple copies (candidate routes) with slight variations in sequencing and resource allocation. These candidate copies capture the essence of optimal resource allocation while allowing quick selection based on real-time conditions, thus achieving both efficiency and speed.
Data Source
AI summary
One challenge for middle-mile route planning is that the set of loads changes significantly between daily planning and execution. Systems and methods are provided for optimizing a transportation plan for a transportation network based on these load changes. The disclosed system re-optimizes a solution by starting from a previously existing plan and previously generated columns (e.g., candidate routes). The disclosed techniques significantly improve the compute time of the system to generate transportation plans that are optimized accordingly to an optimization parameter. The system takes into account the current execution status associated with a given entry of the plan to determine whether the entry should be re-optimized. Entries corresponding to tours that have already commenced, may be at least partially ignored for re-optimization consideration. The disclosed techniques enable state-aware, adaptive re-optimization for even tours that are in-progress or have been tendered.