Flat Sheet Trim Planning With Knapsack-Based Loss Minimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current trimming algorithms in the flat sheet industry primarily focus on minimizing the number of master rolls used to fulfill customer demands, leading to suboptimal solutions and practical unusability, as they do not effectively address trim loss and yield optimization.
Innovation Solution
A method and system that utilize analytics and a Knapsack algorithm to generate optimal trimming patterns, considering trade constraints, machine specifications, and inventory details, with an objective function that minimizes trim loss, integrating upper and lower bound demand constraints, and employing shadow prices to suggest efficient order quantities and sizes for improved operational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional trim algorithms focus on minimizing the number of master rolls used, then the number of rolls is reduced, but trim loss increases and solutions become suboptimal
Solution Approach 1:
The patent changes the optimization parameter from minimizing the number of master rolls to minimizing trim loss. This is achieved by reformulating the objective function in the linear programming model to directly minimize the amount of wasted material rather than the number of rolls used, fundamentally altering the optimization criterion to align with actual production efficiency goals
Solution Approach 2:
Instead of minimizing the number of rolls (traditional approach), the patent inverts the approach by maximizing material utilization or minimizing the waste portion. The optimization model is constructed to minimize trim loss directly, which is the complement of maximizing useful material extraction from each master roll
2Adaptability or versatility
If multiple trimming steps are used to handle different dimensions, then dimension flexibility is improved, but process complexity increases
Solution Approach 1:
The patent creates a universal trim planning system that can handle multiple dimension requirements within a single integrated optimization framework. The linear programming model incorporates various dimension constraints and customer order specifications, allowing the system to determine optimal trimming patterns for different dimensions without requiring separate processing steps for each dimension type
Solution Approach 2:
The patent segments the trimming problem into manageable components (customer orders, master rolls, trimming patterns) that can be independently analyzed and optimized. By breaking down the complex multi-dimensional trimming problem into discrete order fulfillment units with specific width and diameter requirements, the system can optimize each segment while maintaining overall process efficiency
3Loss of substance
If manual planning iterations are performed to optimize trim loss, then trim efficiency can be improved, but time consumption and subjectivity increase
Solution Approach 1:
The patent replaces manual mechanical planning iterations with an automated computational optimization system. The linear programming model and algorithm automatically calculate optimal trimming patterns without requiring manual intervention, eliminating the time-consuming iterative process while providing objective, reproducible results based on mathematical optimization rather than subjective judgment
Solution Approach 2:
The trim planning system is self-sufficient in generating optimal solutions without requiring external manual planning input. The algorithm autonomously processes customer orders, master roll specifications, and constraints to produce optimized trimming patterns, making the system self-service capable and eliminating dependency on manual planning expertise and time investment
Data Source
Figure 1A~1B
Figure 2A~2B
Figure 3~4
AI summary
Planner insight analytics identifies orders and sizes customers can execute to achieve business and operational efficiency. Generating optimal trimming patterns for trimming raw rolls and/or sheets of a flat sheet stock/customer orders includes: (a) receiving customer orders, primary and secondary machine specifications, warehouse inventory, and trade constraints; and (b) generating solutions for (i) order quantity fulfillment, (ii) a primary cutting pattern for the primary machine, (iii) a secondary cutting pattern for the secondary machine, and (iv) inventory details, wherein the solutions are generated with consideration of the initial trade constraints; (c) executing a batch and generating suggestions; (d) generating modified solutions for the parameters in step (b) using revised trade constraints derived from the suggestions generated that override the initial trade constraints in (c); and (e) operating a cutting apparatus. Trim optimization knapsack algorithm with objective function with term relating to trim loss which is eventually minimized is employed.