Surface Mount Production Planning With Grouped Component Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Generating an optimal production plan for a production line is challenging due to the complexity of combinatorial optimization problems involving numerous variables, which can lead to long processing times or combinatorial explosion, resulting in inefficient production plans.
Innovation Solution
A production plan generating device and method that optimizes component allocation and production order using integer programming techniques, such as the GNU linear programming kit (GLPK), to minimize production time by grouping figure numbers and re-optimizing component allocation for each group, thereby reducing the process load and production time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If combinatorial optimization is performed with all variables related to figure numbers and production line, then optimal production plan is achieved, but processing time becomes excessively long or combinatorial explosion occurs
Solution Approach 1:
The patent divides the production plan into multiple blocks, where each block contains a subset of figure numbers. Component allocation is optimized independently for each block rather than for the entire production plan at once. This segmentation reduces the combinatorial complexity from O(n!) for n figure numbers to multiple smaller optimizations, preventing combinatorial explosion while maintaining near-optimal solutions.
Solution Approach 2:
The patent extracts and fixes certain variables related to figure numbers as constants before performing component allocation optimization. By determining fixed figure number assignments in advance and removing them from the optimization variables, the search space is significantly reduced, allowing the optimization to complete in realistic time frames while still achieving optimal component allocation for the remaining variables.
2Productivity
If component allocation is optimized for all figure numbers simultaneously, then production efficiency is maximized, but process load becomes unmanageably high
Solution Approach 1:
The production plan is segmented into multiple blocks, and component allocation optimization is performed separately for each block. This divides the complex global optimization problem into multiple simpler local optimization problems, reducing the computational process load while maintaining overall production efficiency through coordinated block scheduling.
Solution Approach 2:
Certain figure number assignments and production sequences are determined in advance as fixed constants before the component allocation optimization is performed. This preliminary action reduces the number of variables that need to be optimized simultaneously, lowering the process load while still enabling efficient production through pre-planned sequencing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In order to reduce the processing load when generating a production plan for surface mounting lines, a first component allocation determination unit (36) optimizes component allocation for each figure number, with a plurality of variations, and a grouping unit (38) executes, for each component allocation, processing for grouping a plurality of the figure numbers which can be produced without exchanging carriages set for a plurality of mounters. Furthermore, a production order determination unit (42) executes, for each component allocation, processing for optimizing the production order of the grouped figure number groups so as to achieve the shortest production time for all products. Moreover, a production plan determination unit (44) determines, as the production plan for the surface mounting lines, the production plan among the production order optimization results which has the shortest production time.