Hierarchical Decomposition Heuristic for Planning Scheduling Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for planning and scheduling in processing facilities face challenges in finding globally optimal solutions, as they are often computationally intensive and require restrictive assumptions, making it difficult to achieve feasible solutions that satisfy due-dates and deadlines.
Innovation Solution
A hierarchical decomposition heuristic (HDH) is introduced, which decomposes complex decision-making problems into a coordination layer and a cooperation/collaboration layer, using coordinators and cooperators to generate and refine offers based on models, allowing for the identification of globally feasible solutions through iterative communication and model adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If globally optimal solutions are pursued through integration of planning and scheduling problems, then solution optimality is improved, but computational complexity and solution time increase significantly
Solution Approach 1:
The patent divides the integrated planning and scheduling problem into separate hierarchical levels (planning level and scheduling level), allowing each to be solved independently with appropriate granularity. The planning level handles high-level production quantities and due dates, while the scheduling level handles detailed operation sequencing, thereby reducing computational complexity while maintaining solution quality.
Solution Approach 2:
The patent introduces a hierarchical dimension to the problem-solving approach, organizing decisions across multiple levels (strategic planning, tactical scheduling, operational control). This dimensional transformation allows complex integrated problems to be decomposed into manageable sub-problems that can be solved more efficiently at each level.
2Measurement precision
If globally optimal solutions are pursued through integration of planning and scheduling problems, then solution optimality is improved, but device complexity increases
Solution Approach 1:
The system is segmented into distinct functional modules operating at different hierarchical levels. The planning module handles aggregate production decisions, while the scheduling module handles detailed operation sequencing. This segmentation reduces system complexity by creating independent, manageable components with well-defined interfaces.
Solution Approach 2:
The patent introduces intermediate variables and interfaces between planning and scheduling levels that facilitate coordination without requiring full integration. These intermediaries (such as production quantities, due dates, and resource availability) enable information exchange while maintaining modular system architecture.
3Reliability
If restrictive assumptions are made to generate globally optimal solutions, then solution feasibility is improved, but adaptability decreases
Solution Approach 1:
The patent applies different levels of detail and assumptions appropriate to each hierarchical level. At the planning level, aggregate assumptions are used that are less restrictive, while at the scheduling level, more detailed local constraints are applied. This allows the system to maintain feasibility at each level without requiring restrictive global assumptions.
Data Source
AI summary
A system includes a coordinator configured to generate a first offer representing a possible solution to a problem being solved. The coordinator is configured to generate the first offer using a first model. The system also includes multiple cooperators each configured to receive the first offer and to determine if a sub-problem associated with the problem can be solved based on the first offer and using a second model. Each cooperator is also configured to identify an infeasibility and communicate the infeasibility to the coordinator when the sub-problem cannot be solved based on the first offer. The coordinator is further configured to receive the infeasibility, generate a second offer by reducing a deviation between the first offer and a consensus of the cooperators, and provide the second offer to the cooperators. The consensus of the cooperators is based on a level of the first offer that each cooperator can achieve.