Work Process Scheduling Using Candidate Task Scenario Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing scheduling instances of work processes are labor-intensive, infeasible due to combinatorial explosion, or suffer from the curse of dimensionality, making them inefficient and time-consuming.
Innovation Solution
A method that generates a pool of candidate tasks using critical path analysis or similarity-based approaches, followed by a modified greedy algorithm to iteratively generate optimized scheduling scenarios, reducing computation time and improving key performance indicators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional optimization methods are used to generate scheduling scenarios, then scheduling quality can be improved, but computation time increases exponentially due to combinatorial explosion
Solution Approach 1:
The patent segments the scheduling problem by dividing T total tasks into C candidate tasks (where C < T) that form the pool P. This segmentation reduces the combinatorial search space from considering all T tasks to only C critical tasks, thereby reducing computation time while maintaining scheduling quality through focused optimization on the most impactful tasks.
Solution Approach 2:
The patent extracts the C most critical tasks from the full set of T tasks to form the candidate pool P. This extraction is based on identifying tasks that have the greatest impact on scheduling performance, allowing the optimization to focus only on these extracted tasks rather than all tasks, thus avoiding combinatorial explosion while preserving scheduling quality.
2Measurement precision
If exhaustive search methods are applied to evaluate all possible scheduling scenarios, then optimal scheduling can be achieved, but the computational complexity becomes infeasible
Solution Approach 1:
Instead of performing exhaustive search over all T tasks, the patent applies partial action by evaluating only C candidate tasks (where C < T) that are selected to have the most significant impact on scheduling performance. This partial evaluation achieves sufficient optimization accuracy without the infeasible computational complexity of exhaustive search.
Solution Approach 2:
The patent changes the parameter of task selection by introducing a candidate pool size C that is smaller than the total task count T. This parameter change transforms the problem from evaluating all T tasks to evaluating only C critical tasks, reducing computational complexity while maintaining optimization accuracy through focused selection of the most impactful tasks.
3Productivity
If manual scheduling optimization is performed to improve key performance indicators, then scheduling performance can be enhanced, but the process becomes labor-intensive and time-consuming
Solution Approach 1:
The patent implements self-service by providing an automated system that generates optimized scheduling scenarios without requiring manual intervention. The system automatically identifies C candidate tasks, generates k scheduling scenarios, and selects the optimal schedule based on key performance indicators, thereby enhancing productivity while maintaining operational simplicity through automation.
Solution Approach 2:
The patent replaces manual mechanical scheduling operations with an automated computational system. Instead of labor-intensive manual optimization, the system uses automated algorithms to generate and evaluate scheduling scenarios, substituting human effort with machine-based optimization that improves productivity while keeping the process simple to operate.
Data Source
AI summary
A method, computer program product, and computer system optimizing a scheduling instance for a work process and performing the work process in accordance with the optimized scheduling instance. A pool P of C candidate tasks of a work process is generated. The work process is characterized by an initial scheduling instance that includes tasks. A total of k output scenarios are generated, where k≥2. Each output scenario includes b tasks selected from the pool P, where b≥2. A modified schedule instance (MSI) for each output scenario is generated and key performance indicators (KPIs) associated with each MSI are generated. The generated MSIs and KPIs are outputted. One MSI of the generated MSIs is selected. The work process is performed in accordance with the selected one MSI, wherein the KPIs associated with the selected one MSI improve the work process.