Vehicle Production Sequencing With Interaction-Model Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods lack a systematic approach to compute a production plan and sequence satisfying constraint conditions for multi-product mixed production in a motor vehicle production line, where vehicles of multiple types or specifications are produced simultaneously.
Innovation Solution
A system utilizing an interaction model, where the computational device computes a planned sequence of production based on management information, assuming assignment events of sequential positions in the production process as variables and setting constraint conditions as interaction strengths between variables.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sequential execution of instructions is used on Von Neumann architecture computer, then basic operation is simple, but processing speed and scale of soluble problem are restricted due to operating clock frequency limits
Solution Approach 1:
The patent replaces the sequential mechanical instruction execution of Von Neumann architecture with quantum mechanical processes. Quantum computers utilize quantum bits (qubits) that can exist in superposition states and perform parallel computations through quantum interference, fundamentally substituting the classical mechanical computation model with quantum physical processes to achieve exponential speedup for certain optimization problems
Solution Approach 2:
The patent changes the fundamental parameters of computation by transitioning from classical binary states (0 or 1) to quantum states that can represent multiple values simultaneously through superposition. This parameter change allows the system to evaluate multiple solution candidates in parallel rather than sequentially, dramatically improving processing speed for combinatorial optimization problems
2Measurement precision
If heuristic method is used to solve large-scale combinatorial optimization problem, then solution can be found, but processing speed is limited and may not reach optimal solution
Solution Approach 1:
The patent employs simulated annealing which utilizes phase transition concepts from thermodynamics. The system starts with high 'temperature' allowing random transitions between states to escape local minima, then gradually reduces temperature to converge toward the global optimum. This phase transition approach balances exploration and exploitation, achieving high optimization accuracy while controlling processing time through the annealing schedule
Solution Approach 2:
The patent implements an Ising model that can represent various combinatorial optimization problems in a unified framework. By mapping different problem types onto the same Ising Hamiltonian structure, the system achieves universal applicability across multiple optimization domains, allowing a single computational approach to solve diverse problems with both accuracy and efficiency
3Adaptability or versatility
If multi-product mixed production is performed with vehicles of multiple types or specifications, then production flexibility is improved, but production planning complexity increases due to varying working loads and constraint conditions
Solution Approach 1:
The patent segments the production planning problem into discrete assignment variables for each vehicle and time slot, with binary decision variables indicating whether a vehicle is assigned to a specific position. This segmentation transforms the complex continuous planning problem into discrete combinatorial optimization that can be solved using quantum computing methods, managing complexity through structured decomposition
Solution Approach 2:
The patent introduces an interaction model as an intermediary between the production constraints and the optimization algorithm. The interaction model encodes constraint conditions (such as working load balances, sequence requirements, and resource availability) into an energy landscape where valid solutions correspond to low-energy states. This intermediary representation simplifies the complex constraint satisfaction problem into a form suitable for quantum annealing optimization
Data Source
AI summary
Computation is made of a production plan and sequence satisfying constraint conditions using an interaction model. A system includes a storage device to store management information about specifications on each of a plurality of things to be produced and a computational device to compute a planned sequence of things to be produced in a production process to produce the plurality of things using an interaction model, based on the management information. In this system, the computational device computes an interaction model in which an assignment event of a sequential position in a production process to each of the things is assumed as a variable and a constraint condition regarding the production process is set as strength of an interaction between variables.