Mixed-Fleet Route Sequencing Under Dynamic Traffic Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current GPS navigation systems fail to dynamically and optimally sequence multi-destination routes due to their inability to adapt to dynamic traffic patterns and real-world conditions, leading to inefficient routes and potential traffic congestion.
Innovation Solution
A system and method that utilize real-time, historical, and predicted traffic data, combined with machine learning and AI, to continually re-sequence routes based on various constraints such as distance, time, and traffic conditions, while integrating sensors and data from multiple sources to provide optimal route optimization for single and multiple vehicle fleets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If GPS navigation systems treat all traffic conditions as equal and output routes in the exact sequence inputted, then the navigation system is simple to operate, but the route efficiency deteriorates and time is lost in traffic
Solution Approach 1:
The patent implements dynamic route sequencing that automatically reorders destinations based on real-time traffic conditions, vehicle positions, and predictive analytics. The system continuously monitors traffic patterns and recalculates the optimal visit sequence, transforming the static input-order routing into a dynamic, adaptive process that minimizes travel time while maintaining ease of use through automated decision-making
Solution Approach 2:
The system incorporates real-time feedback loops that monitor actual traffic conditions, vehicle locations, and arrival times. This feedback is fed back into the routing algorithm to continuously refine and adjust the destination sequence, enabling the system to adapt to changing traffic patterns and optimize routes dynamically without requiring user intervention
2Object-affected harmful factors
If vehicles are re-routed to avoid congested areas, then traffic congestion is reduced, but vehicles may converge in other areas and cause new congestion
Solution Approach 1:
The patent combines fleet-wide routing data and predictive analytics to coordinate route adjustments across multiple vehicles. By merging individual vehicle routing decisions into a unified fleet optimization system, the patent distributes traffic flow more evenly across the network, preventing vehicles from independently converging on the same alternative routes and creating new congestion points
Solution Approach 2:
The system performs preliminary routing adjustments by predicting future traffic patterns and proactively redistributing vehicles before congestion occurs. By anticipating potential convergence points and adjusting routes in advance, the system prevents the formation of new congestion areas rather than reacting after vehicles have already gathered
3Productivity
If dynamic route re-sequencing is implemented based on real-time traffic data, then route efficiency is improved and time is saved, but system complexity increases
Solution Approach 1:
The patent implements self-service routing where the system automatically monitors traffic conditions, calculates optimal sequences, and adjusts routes without requiring user input or intervention. The navigation system serves itself by making intelligent routing decisions based on real-time data, thereby improving efficiency while maintaining a simple user interface that does not expose the underlying complexity
Data Source
AI summary
Methods and systems to dynamically and optimally sequence routes based on historical and real-time traffic conditions, and to predict anticipated traffic conditions along the dynamically generated route are disclosed. Route sequencing may be based on a set of predefined constraints, e.g., distance, time, time with traffic, or any objective cost function. The system of the present invention may be implemented in a vehicle fleet comprising one or more vehicles with one or more depots, or with no depots. An optimization server obtains real-time, historical and/or predicted future traffic, weather, hazard, and avoidance-zone data on road segments to generate a route, while staying within parameters and constraints set by an automatic machine learning process, an artificial intelligence program, or a human administrator. The platform may be coupled to sensors positioned on roads, e.g., speed radar or camera, and sensors positioned in vehicles, e.g., GPS system or on-board diagnostic sensor.