Block-Based Constraint Workflows for Semiconductor Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing constraint programming techniques in manufacturing environments require custom code, making them inflexible and difficult to maintain, especially for end-users without programming expertise, which hinders adaptation to changes in manufacturing processes and increases costs.
Innovation Solution
Implementing block-based workflows that allow end-users to construct and modify constraint programming models without coding, using a series of blocks to specify operations, extract data, create models, and solve constraints, enabling flexible and extensible scheduling systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If constraint programming models are created using custom code, then the system can solve scheduling problems effectively, but the system becomes difficult to maintain and inflexible when manufacturing conditions change
Solution Approach 1:
The constraint programming model is segmented into discrete, configurable parameters and constraints that can be individually modified. Instead of a monolithic code structure, the system divides the scheduling model into separate components (objective functions, constraints, variables) that can be adjusted independently when manufacturing conditions change, enabling flexible adaptation without rewriting the entire system.
Solution Approach 2:
The system implements dynamic configurability where scheduling parameters, constraints, and model structures can be changed in response to varying manufacturing conditions. The model transitions from a static code-based approach to a dynamic system that can adapt its parameters and constraints based on current production requirements, equipment availability, and changing priorities.
2Adaptability or versatility
If constraint programming models use custom code, then functional requirements can be met, but technical expertise and time commitment increase significantly
Solution Approach 1:
The system uses template-based model configurations that can be copied and reused across different scheduling scenarios. Pre-defined constraint patterns, objective function templates, and model structures serve as reusable copies that can be instantiated with different parameters, reducing the need to create complex code from scratch and lowering the barrier to entry for users without programming expertise.
Solution Approach 2:
The constraint programming framework implements universal, parameter-driven models that can handle multiple scheduling scenarios and manufacturing conditions through a single configurable system. Rather than requiring separate code implementations for different scheduling problems, the system uses a unified model structure that can be adapted to various functional requirements through parameter adjustment, reducing overall system complexity.
3Productivity
If manufacturing facilities implement scheduling systems to manage complexity, then productivity improves, but implementation and maintenance costs increase
Solution Approach 1:
The system enables productivity improvements through parameter-driven scheduling models where key performance parameters (production rates, cycle times, priority weights) can be adjusted to optimize throughput. By focusing on parameter optimization rather than structural code changes, the system achieves productivity gains while minimizing implementation complexity and maintenance costs.
Data Source
AI summary
Embodiments presented herein provide techniques for executing a block-based (BB) workflow to solve a constraint programming (CP) model related to a semiconductor manufacturing environment. Embodiments include receiving at least one BB workflow comprising a plurality of blocks. The plurality of blocks may specify a set of operations. Embodiments include accessing a plurality of block definitions corresponding to the plurality of blocks. Embodiments include executing the at least one BB workflow by performing the set of operations based on the plurality of block definitions, including extracting data from the semiconductor manufacturing environment, the data comprising both static data and dynamic data related to equipment in the manufacturing environment, creating the CP model based on the data and at least one constraint defined in the BB workflow, using a solver to determine a solution to the CP model; and publishing the solution to at least one component in the semiconductor manufacturing environment.