Autonomous Document Conveyance With Failure-Rate Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional document processing apparatuses face challenges in suppressing conveyance failures and maintaining productivity due to insufficient examination of reading condition combinations, requiring manual input of teacher data and limited freedom in document processing information, leading to inefficiencies and increased effort.
Innovation Solution
A document processing apparatus with autonomous conveyance control that estimates conveyance-failure occurrence rates based on combination data, allowing for autonomous conveyance and control adjustments without external instruction, using a conveyor, reading device, and circuitry to determine optimal conveyance parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual input of teacher data is used for machine learning, then conveyance failure can be detected, but the work for acquiring combination data requires time and effort
Solution Approach 1:
The document processing apparatus autonomously feeds and conveys documents without external instructions, automatically acquiring combination data of document processing information and conveyance failure occurrence probabilities. This self-service mechanism eliminates manual data input operations, reducing the time and effort required for teacher data acquisition while maintaining conveyance failure detection capability
Solution Approach 2:
The system performs preliminary autonomous conveyance operations to collect combination data before actual document processing. By pre-acquiring teacher data through autonomous mode operations, the system prepares the machine learning model in advance, eliminating the need for manual data collection during production operations
2Reliability
If the number of reading condition combinations is increased to cover all scenarios, then conveyance failure can be suppressed, but the design examination workload increases enormously
Solution Approach 1:
The document processing apparatus autonomously explores different reading condition combinations and document types during autonomous conveyance mode, automatically generating comprehensive combination data without requiring exhaustive manual design examinations. This self-service data collection approach covers all scenarios systematically while keeping design workload manageable
Solution Approach 2:
The system dynamically adjusts reading conditions and document parameters during autonomous operations to explore the solution space. By making the data collection process dynamic and adaptive rather than static and exhaustive, the system efficiently covers all scenarios without requiring predetermined examination of every combination
3Reliability
If sheet feeding speed is decreased to prevent conveyance failure, then conveyance reliability improves, but productivity decreases
Solution Approach 1:
The machine learning model determines optimal sheet feeding speeds by analyzing combination data that includes conveyance failure occurrence probabilities at different speeds. This parameter optimization approach adjusts the sheet feeding speed to the precise value needed for reliable conveyance without unnecessarily reducing it, thereby maintaining both conveyance reliability and productivity
Solution Approach 2:
The system uses feedback from autonomous conveyance operations to continuously refine the relationship between sheet feeding speed and conveyance failure probability. By incorporating this feedback into the machine learning model, the system learns the optimal speed settings that prevent failures while maximizing productivity
4Ease of operation
If autonomous conveyance mode is implemented to reduce manual effort, then data acquisition efficiency improves, but the degree of freedom in document processing information is limited
Solution Approach 1:
The autonomous conveyance mode is designed to handle multiple document types and processing conditions universally. By implementing a versatile autonomous mode that can adapt to various document characteristics and processing requirements, the system maintains high data acquisition efficiency while preserving the degree of freedom in document processing information
Data Source
AI summary
A document processing apparatus includes a document tray, a conveyor, a reading device, and circuitry. The conveyor conveys an original document from the document tray. The reading device reads an image on the original document conveyed by the conveyor as a reading operation. The circuitry is to acquire document processing information of the original document, estimate a conveyance-failure occurrence rate when the reading device reads the image on the original document while the conveyor conveys the original document, based on combination data of the document processing information and the conveyance-failure occurrence rate when the conveyor autonomously conveys the original document in an autonomous conveyance mode without an instruction from an image forming apparatus, and determine a conveyance control amount of the conveyor to convey the original document according to the conveyance-failure occurrence rate estimated.