Machine Tool Workstation Sequencing for High-Mix Setup Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manufacturing industries face significant challenges with long setup times in machine tool workstations, leading to increased inventory costs, reduced efficiency, and poor on-time delivery due to the need for frequent changes in part types and processes, which existing methods like Lean Six Sigma and heuristics fail to effectively address, especially in high-mix, low-volume production environments.
Innovation Solution
Implementing artificial intelligence and machine learning techniques to group machine tool workstations by type and train neural networks to optimize part processing sequences, using data from sensors to minimize setup times while ensuring on-time delivery, by dynamically calculating minimum cycle times and adjusting sequences based on Little's Law.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If machine tools are reconfigured for different part types, then product variety is improved, but setup time increases
Solution Approach 1:
The system performs preliminary actions by pre-calculating and storing optimal processing sequences for different part combinations in a database. When new parts arrive, the system quickly retrieves pre-computed sequences rather than calculating from scratch, thereby reducing setup time while maintaining adaptability to different part types.
Solution Approach 2:
The system dynamically adjusts processing sequences based on real-time inputs including part geometry, material properties, and machine tool capabilities. The neural network continuously learns from new data and adapts its recommendations, allowing the system to optimize for both product variety and minimal setup time as conditions change.
2Productivity
If batch size is increased to compensate for setup time, then productivity is improved, but inventory cost increases
Solution Approach 1:
The system changes the parameter of batch size from a fixed large value to a dynamically optimized value. By analyzing part similarity, machine tool capabilities, and setup time requirements, the system determines optimal batch sizes that minimize total cost while maintaining productivity. This allows smaller, more frequent batches when part variety is high, and larger batches when part similarity is high.
3Loss of time
If engineering-intensive methods are used to reduce setup time, then setup time is reduced, but method complexity increases
Solution Approach 1:
The system replaces complex engineering-intensive manual methods with an automated neural network-based information processing system. The neural network automatically analyzes part geometries, material properties, and machine capabilities to generate optimized processing sequences, eliminating the need for manual engineering analysis while reducing setup time.
Solution Approach 2:
The system creates digital copies of part designs, material specifications, and machine tool capabilities in a database. These digital models are used by the neural network to simulate and optimize processing sequences without requiring physical prototypes or manual trial-and-error, thereby reducing setup time and method complexity.
4Adaptability or versatility
If frequent part type changes are implemented, then market responsiveness is improved, but setup frequency increases
Solution Approach 1:
The system merges the processing of multiple different part types into single production runs by identifying and exploiting similarities between parts. The neural network groups parts that can be processed using similar machine tool configurations and parameters, allowing the system to respond to market diversity while minimizing the number of actual setup changes required.
Data Source
AI summary
Methods, systems and apparatus, including computer programs encoded on computer storage medium, for controlling operations of machine tool workstations. Machine tool workstations are grouped into functional groups. Neural networks corresponding to the functional groups are trained to process respective inputs representing parts to be processed to generate respective outputs representing sequences of ordered subsets of the parts that produce a reduced setup time for workstations in the functional groups. Data representing respective collections of parts to be processed by workstations included in the functional groups is processed using the trained neural networks to generate corresponding sequences of ordered subsets of the collection of parts. Average delay times associated with the generated sequences of ordered subsets of the collection of parts are computed. If the average delay times are less than a predetermined threshold, parts are released to the functional groups for processing according to the generated sequences.