Rolling-Horizon Production Planning for Electronics Time Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of planning and scheduling multi-stage production processes in electronics manufacturing is exacerbated by sequence-dependent setup times, varying lot sizes, shared resources, and de-coupling of production stages, leading to inefficiencies and challenges in optimizing production processes.
Innovation Solution
A computer-implemented method and system that iteratively combines optimization and simulation using a rolling horizon approach, where production control parameters are optimized and simulated within time windows, iteratively refining parameters to achieve efficient production planning across a planning time horizon.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detailed simulation and optimization are performed for the entire planning time horizon, then production planning quality is improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The planning time horizon is divided into multiple time windows, with the planning horizon further segmented into time periods. This segmentation allows the optimization and simulation to be performed on smaller, manageable segments rather than the entire time horizon at once, reducing computational complexity while maintaining planning quality through iterative refinement across segments.
Solution Approach 2:
The method performs optimization and detailed simulation iteratively for each time period and time window in a systematic sequence. By preparing and optimizing each segment beforehand and using rolling horizon procedures, the system achieves comprehensive planning quality without requiring all computations to occur simultaneously, thus managing processing time effectively.
2Manufacturing precision
If sequence-dependent setup times are considered in production planning, then production accuracy is improved, but planning complexity increases
Solution Approach 1:
The system dynamically adjusts production control parameters including setup times based on the specific sequence of products being manufactured. By incorporating sequence-dependent setup times as variable parameters in the optimization model rather than fixed values, the system achieves higher production accuracy while managing complexity through parameter-based modeling within the iterative optimization framework.
3Productivity
If varying lot sizes are optimized for different products and stages, then production efficiency is improved, but planning complexity increases
Solution Approach 1:
The system optimizes lot sizes locally for each intermediate product and production stage rather than applying uniform lot sizes across the entire production process. This allows each production stage and product type to have customized lot sizes tailored to its specific requirements, improving production efficiency while managing complexity through localized optimization within the iterative framework.
4Adaptability or versatility
If production stages are de-coupled with supermarkets, then production flexibility is improved, but inventory requirements increase
Solution Approach 1:
The system dynamically manages inventory levels in supermarkets (intermediate storage points between de-coupled production stages) based on actual production progress and demand signals. By making inventory levels dynamic rather than static, the system achieves production flexibility through de-coupling while minimizing inventory requirements through real-time adjustment of stock levels in response to changing conditions.
Data Source
Figure 1~4
AI summary
The invention relates to a computer-implemented method for planning a production process of electronics products within a time period of a sequence of time windows within a planning time horizon, where the time period recurs at time intervals of a pre-defined length in a rolling horizon procedure until the entire planning time horizon has been planned, whereby the method comprises the following steps: a) for every time period an optimization (OPT) uses initial input planning parameters and determines (03) production control parameters for each current time window in the current time period and hands them over (04) to a simulation (SIM) b) which uses these production control parameters and simulates a course of the production process in detail (05) and which provides simulation results of the simulated course of the production process in the sequence of the first time windows within the time period, wherein these results are passed back (02) to the optimization (OPT) c) which is repeated using updated input planning parameters, which are at least partly derived from said simulation results, for the next time interval that shifts the time period by a time interval of the pre-defined length of the rolling horizon procedure, and d) steps a), b) and c) are repeated until the whole planning time horizon is simulated.