Abstract:The assembly takes up a special position in a company's order processing; as with regard to the company's service provision, it mostly comes at the end of the value chain. For that reason the assembly depicts a melting pot of many technical, organisational and qualitative errors. To mitigate technical failure, in the past versatile solutions were developed, but there are only little satisfactory solutions to consider human failure in the planning process of manual assembly operations. This paper presents a procedure that allows transferring the knowledge about human reliability gained in safety-critical industries to manual assembly operations. In this context, at the research institute of the authors, approaches have been developed to use the knowledge of the Expert System for Task Taxonomy – a method that originally was developed to evaluate error rates of control activities in safety-critical areas – for creating a new method that allows evaluating human reliabilities in manual assembly operations. The aim of this method is to be able to systematically predict human error probabilities for any manual work content of the industrial batch production. Considerable elements of the developed assembly-specific method are an own database containing assembly-specific standard terms and a calculation model which is adapted to low error rates of the manual assembly. First results at a heating manufacturer verified that the developed method enables reliable risk predictions for the field of manual assembly. This is particularly given by the creation of a new database, which is especially tailored to the application field of manual assembly and the development of assembly-specific calculation regulations. In the following steps, the comparison of the calculated error data and the actual error data of the industrial practice is extended, so that the method can be validated across all industries.

Using Two-Level Context-Based Predictors for Assembly Assistance in Smart Factories

Assembly Assistance System with Decision Trees and Ensemble Learning

Robust Assembly Assistance Using Informed Tree Search with Markov Chains

Learning-Based Hand Motion Capture and Understanding in Assembly Process

Prediction-Based Human-Robot Collaboration in Assembly Tasks Using a Learning from Demonstration Model

Hand-in-Hand Guidance: an Explore-Exploit Based Reinforcement Learning Method for Performance Driven Assembly-Adjustment

A Digital Twin-Driven Guidance Method for Human-Machine Collaborative Assembly Operations Based on Machine Learning and Computer Vision

Context-Aware Human Activity Recognition in Industrial Processes

Two-Stage Clustering of Human Preferences for Action Prediction in Assembly Tasks

A Comparative Study of Digital Assembly Assistance Systems

Toward Proactive Human–Robot Collaborative Assembly: A Multimodal Transfer-Learning-Enabled Action Prediction Approach

Multiscale Cost-Sensitive Learning-Based Assembly Quality Prediction Approach under Imbalanced Data

An Integrated Method for High-Dimensional Imbalanced Assembly Quality Prediction Supported by Edge Computing

Predicting Assembly Geometric Errors Based on Transformer Neural Networks

Augmented Reality Assisted Assembly Training Oriented Dynamic Gesture Recognition and Prediction

A graph-based reinforcement learning-enabled approach for adaptive human-robot collaborative assembly operations

Assembly-specific database for predicting human reliability in assembly operations

SHARIDEAS: a smart collaborative assembly platform based on augmented reality supporting assembly intention recognition

A flexible manufacturing assembly system with deep reinforcement learning

A reinforcement learning method for human-robot collaboration in assembly tasks

Detection and Estimation of Mental Fatigue in Manual Assembly Process of Complex Products