RIP Setting Optimization via Machine Learning for Print Speed

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Determining optimal RIP setting conditions to maximize printing speed is challenging due to the numerous combinations of data sizes and print setting conditions, making it difficult to create conditional expressions that result in the highest printing speed.

Innovation Solution

A print data generating apparatus that employs a machine learning unit to predict printing speeds based on data size and print setting conditions, and adjusts RIP setting conditions to match the processing speed of the printing unit, using a preflight unit to analyze print jobs and a RIP setting condition determining unit to optimize thread allocation and memory usage for efficient RIP processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional conditional expressions are used to determine RIP settings, then the system is simple to implement, but it cannot handle the countless combinations of data sizes and print setting conditions to maximize printing speed

Engineering Contradiction:
Improveprinting speedVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system performs self-learning by automatically collecting actual printing speeds under various conditions and using this data to improve its own prediction accuracy. The machine learning unit continuously refines its models based on real-world performance data, enabling the system to optimize RIP settings autonomously without requiring manual intervention or complex pre-programmed conditional logic for every possible scenario.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system dynamically changes RIP processing parameters (such as number of threads, band divisions, and memory allocation) based on predicted printing speeds for different conditions. By adjusting these parameters according to the specific data size and print settings, the system optimizes printing speed for each unique combination without requiring exhaustive conditional expressions.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If the number of RIP processing threads is increased to improve printing speed, then processing capacity increases, but memory consumption and system resource usage increase

Engineering Contradiction:
ImproveRIP processing speedVSAvoidmemory consumption
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The system dynamically adjusts the number of RIP processing threads and other resource allocations based on the predicted printing speed for the specific job conditions. Rather than using a fixed or maximum thread count, the machine learning model determines the optimal thread number that balances processing speed with memory consumption, allowing the system to adapt resource usage to actual needs.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes RIP processing parameters including thread count, band divisions, and memory allocation based on the relationship between data size, print settings, and predicted printing speed. This allows optimal resource utilization where more threads are used when beneficial for speed, while reducing thread count and memory usage when it would be wasteful.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3745244B1Print data generating apparatus
Publication Date: 2023.08.16 RISO KAGAKU CORP
  • EP3745244B1 patent drawingFigure 1
  • EP3745244B1 patent drawingFigure 2
  • EP3745244B1 patent drawingFigure 3

AI summary

To provide a print data generating apparatus that determines RIP setting conditions which are optimized such that printing speed is maximized. A print data generating apparatus includes a machine learning unit 12 that has a learned model which is obtained by inputting and conducting machine learning of the relationships among data size of document data of a plurality of print jobs, print setting conditions, RIP setting conditions and printing speeds when the document data are printed by a printing unit; a printing speed predicting unit 13 that predicts the printing speed of a print job to be printed by inputting the data size of the document data of the print job to be printed, the print setting conditions, and arbitrary RIP setting conditions into the learned model; an optimized RIP setting condition determining unit 14 that determines optimized RIP setting conditions that result in the printing speed which is predicted by the printing speed predicting unit 13 approaching or becoming greater than or equal to a processing seed of the printing unit, and a RIP processing unit 15 that performs a RIP process for the print job to be printed based on the optimized RIP setting conditions which are determined by the RIP setting condition determining unit 14 to generate print data, and outputs the generated print data to the printing unit.