Rail Conflict Resolution Using Genetic Algorithm Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for resolving conflict situations in rail-bound transportation systems require significant computing time and power to find suitable solutions, as they consider all possible combinations of measures, leading to inefficiencies in solving both the original and subsequent conflicts.
Innovation Solution
A method that utilizes a decision support system combining deterministic and heuristic approaches, employing a genetic algorithm to optimize solutions from a reduced action space based on rated solutions from a knowledge base, and applying domain rules to quickly find and implement effective conflict resolutions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all possible combinations of measures are simulated to find a suitable conflict resolution solution, then the completeness and reliability of the solution is improved, but the computing time and power required increases significantly
Solution Approach 1:
The patent segments the solution space by classifying conflict situations into different conflict classes based on operation states. Instead of evaluating all possible measure combinations, the system divides the problem into manageable segments (conflict classes) and pre-determines suitable measures for each class, storing them in a knowledge base. This segmentation reduces the computational burden while maintaining solution reliability.
Solution Approach 2:
The system performs preliminary action by pre-simulating and rating multiple measure combinations for each conflict class before actual conflict resolution is needed. The results are stored in a knowledge base with pre-calculated ratings. When a conflict occurs, the system quickly retrieves and applies pre-evaluated solutions rather than performing full simulations in real-time, significantly reducing computing time while maintaining solution quality.
2Productivity
If a knowledge base with pre-rated solutions is used to reduce computing time, then the speed of conflict resolution is improved, but the adaptability to unique or novel conflict situations may be reduced
Solution Approach 1:
The system incorporates feedback mechanisms where the actual outcomes of applied measures are fed back into the knowledge base to update and refine the ratings of different measure combinations. This continuous learning process allows the system to adapt to novel conflict situations by improving its knowledge base over time, balancing the use of pre-rated solutions with the ability to handle unique scenarios effectively.
3Device complexity
If the action space is reduced by assigning measures to conflict classes, then the complexity of the decision-making process is reduced, but the precision of matching the optimal measure to the actual situation may be reduced
Solution Approach 1:
The system changes parameters by classifying conflict situations based on key operation state parameters and assigning them to conflict classes. This parameter-based classification simplifies the decision-making process by reducing the vast action space into manageable categories. The knowledge base stores pre-evaluated measures for each class, enabling quick retrieval while maintaining sufficient matching precision through the structured classification approach.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention concerns a method for controlling vehicles in case of a conflict situation, with the following steps: (a) determination of an actual operation state of the vehicle and/or of a planned route; (b) classification of the conflict situation thereby determining an action space; (c) searching a knowledge base comprising rated solutions which are assigned to an operation state equal or similar to the actual operation state; in case one or more rated solutions are found which are assigned to an operation state which is similar to the actual operation state: (d) selection of at least one rated solution from the knowledge base; (e) optimization of the at least one selected solution applied to the actual operation state, wherein the optimization uses a genetic algorithm and results in an optimized solution; (f) carrying out the optimized solution. Suitable solutions to solve the conflict can be found with reduced time effort.