Print Job Scheduling Using Process Mining for Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In print shops, a significant portion of jobs take longer than expected, affecting overall productivity due to random resource allocation and inefficient printer selection, without considering operator experience or printer capabilities.
Innovation Solution
Implement process mining techniques to collect and analyze past job log data, using machine learning to optimize resource assignment and provide real-time recommendations for improving job completion times and productivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If operators and printers are assigned randomly without considering experience or efficiency, then resource allocation is simple and quick, but job completion time increases and productivity decreases
Solution Approach 1:
The system implements feedback by collecting actual job completion data from past print jobs and using this information to continuously improve operator-printer assignments. The process mining algorithm analyzes historical performance data and feeds back optimized assignment recommendations to the scheduling system, creating a closed-loop improvement mechanism that increases productivity while managing complexity through automated data-driven decisions.
Solution Approach 2:
The system performs preliminary action by pre-analyzing operator experience levels, printer capabilities, and historical job performance data before making assignments. The process mining algorithm pre-processes historical data to identify patterns and optimize assignments in advance, so that when jobs need to be scheduled, the system can quickly allocate resources based on pre-computed optimal matches rather than making random assignments.
2Loss of time
If manager predicts job time for each printer and selects shortest time printer, then job completion time may be reduced, but prediction accuracy is insufficient due to lack of detailed operator experience data
Solution Approach 1:
The system applies segmentation by breaking down the job time prediction into distinct components: operator-specific performance data, printer-specific characteristics, job complexity factors, and historical completion data. The process mining algorithm segments historical job data by operator, printer, and job type to identify specific patterns and performance metrics for each combination, enabling more accurate predictions than general manager estimates.
Solution Approach 2:
The process mining algorithm serves as an intermediary between historical data and prediction requirements. It processes raw historical job completion data, extracts relevant performance patterns, and generates accurate time predictions that bridge the gap between available data and managerial decision-making needs, improving prediction accuracy without requiring managers to manually analyze detailed historical records.
3Reliability
If detailed job information is collected and analyzed using process mining, then resource allocation accuracy improves, but data processing complexity and time increase
Solution Approach 1:
The system implements self-service by enabling automated collection, processing, and analysis of job information through the process mining algorithm. The system automatically gathers data from various sources, processes it through mining algorithms to identify patterns and performance metrics, and generates optimized assignment recommendations without requiring manual intervention. This automated self-service approach improves allocation accuracy while managing processing complexity through systematic automated procedures.
Solution Approach 2:
The process mining algorithm serves multiple functions: it collects data from various sources, analyzes historical patterns, generates predictions, and provides optimization recommendations. This multi-functional universal algorithm handles diverse data types and analysis requirements through a single integrated system, improving reliability while containing complexity by consolidating multiple processing functions into one versatile computational engine.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method, a non-transitory computer readable medium, and a system are disclosed for processing and scheduling print jobs on a plurality of printers. The method includes: collecting job information on one or more past print jobs; collecting job information on a plurality of print jobs currently being processed, each of the plurality of print jobs being assigned to one or more operators from a plurality of operators; generating a process workflow visualization for the plurality of print jobs currently being processed with a processing mining algorithm based on the job information collected on the one or more past print jobs and the job information collected on the plurality of print jobs currently being processed; and displaying the process workflow visualization for the plurality of print jobs currently being processed.