Autonomous Production System Decision Transparency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous production systems in Industry 4.0 environments lack transparency in decision-making processes, leading to a lack of control and trust for human operators. There is a need to provide clear explanations for resource choices and production route selections to ensure informed decision-making, especially in cases of non-producibility.

Innovation Solution

A system is developed to provide transparent explanations for production decisions by analyzing alternative production process variants, identifying constraint violations, and presenting detailed information on why certain resources or routes are chosen or rejected. This includes categorizing constraint violations into three levels and providing explanations for valid alternatives and constraint fulfillment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If autonomous production systems make decisions without human intervention, then productivity and automation extent are improved, but transparency and trust for human operators deteriorate

Engineering Contradiction:
Improveproduction efficiencyVSAvoiddecision transparency
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The system implements feedback by providing operators with explanatory information about autonomous decisions. The explanation facility delivers feedback on why certain resources were selected, what alternatives were considered, and what constraints influenced the decision, thereby maintaining transparency while preserving automation benefits

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The explanation facility acts as an intermediary between the autonomous production system and human operators. It translates complex system decisions into comprehensible explanations, bridging the information gap without requiring human intervention in the actual production control

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If the system provides detailed explanations for all production decisions, then transparency is improved, but device complexity and information processing requirements worsen

Engineering Contradiction:
Improvedecision transparencyVSAvoidexplanation system complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The explanation system segments information by categorizing constraints into different levels (Level 1: resource capabilities, Level 2: production rules, Level 3: logistical constraints). This segmentation allows the system to provide detailed explanations without overwhelming complexity, organizing information hierarchically for better manageability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system applies local quality by providing different levels of explanation detail based on the specific decision context. Not all decisions require the same depth of explanation, and the system adapts the level of detail to the particular resource selection or route determination being explained

Inventive Principle:
Principle #3Local quality

3Measurement precision

If the system analyzes all alternative production variants, then manufacturing precision in decision-making is improved, but loss of time in computation increases

Engineering Contradiction:
Improvedecision accuracyVSAvoidcomputation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-establishing the constraint hierarchy and classification framework before actual production decisions are made. This preparatory structuring allows for faster evaluation of alternatives during runtime, as the evaluation criteria are already organized and ready for application

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system manages the trade-off by changing parameters of the analysis process - adjusting which constraints are evaluated at different levels and when. The three-level constraint evaluation structure allows the system to vary the depth of analysis based on decision criticality, balancing accuracy with computation time

Inventive Principle:
Principle #35Parameter changes

4Reliability

If the system provides comprehensive information on constraint violations and alternatives, then reliability of production planning is improved, but ease of operation for operators deteriorates

Engineering Contradiction:
Improveproduction planning reliabilityVSAvoidoperator usability
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The constraint information is segmented into three distinct levels (resource capabilities, production rules, logistical constraints), making comprehensive information more digestible. Operators can understand constraint violations systematically rather than being overwhelmed by undifferentiated information

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The explanation facility serves as an intermediary that translates complex constraint violation data into operator-friendly explanations. It mediates between the detailed system analysis and operator comprehension needs, presenting information in an accessible format while maintaining completeness

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3667576B1Autonomous flexible production system
Publication Date: 2025.03.05 SIEMENS AG
  • EP3667576B1 patent drawingFigure 1~2
  • EP3667576B1 patent drawingFigure 3

AI summary

The invention is related to an autonomous, flexible production system with a set of production machines and a transportation system for carrying out the material flows between the machines in which the production machines can be flexibly connected/disconnected to a production process, said system comprising - means for finding all possible production process variants for a given product that are compatible with the restrictions on the ordering of production operations and involve machines with the right capabilities and appropriate logistical connections between them; - means for characterizing process variants by a chosen allowed ordering of operations and by an allowed assignment of machines to the operations; - means for finding an optimized schedule for the order stack by selecting one process for each order in the stack and setting up a time schedule for all these processes that is optimal with respect to a certain evaluation function, - means for presenting all possible production process variants for a given product or - if there has been no possible production process been found, - means for presenting the possible reasons for unproducibility.