Operating a complex and dynamic system, such as an agile manufacturing line, is a knowledge-intensive task. It imposes a steep learning curve on novice operators and prompts experienced operators to continuously discover new knowledge, share it, and retain it. In practice, training novices is resource-intensive, and the knowledge discovered by experts is not shared effectively. To tackle these challenges, we developed an AI-powered pervasive system that provides cognitive augmentation to users of complex systems. We present an AI cognitive assistant that provides on-the-job training to novices while acquiring and sharing (tacit) knowledge from experts. Cognitive support is provided as dialectic recommendations for standard work instructions, decision-making, training material, and knowledge acquisition. These recommendations are adjusted to the user and context to minimize interruption and maximize relevance. In this article, we describe how we implemented the cognitive assistant, how it interacts with users, its usage scenarios, and the challenges and opportunities.@en

Diversey BV, a major player of professional hygiene product manufacturing, is facing challenges with agile manufacturing of hygienic products with changeover process consuming most of production time. They are collaborating with the EU-Horizon 2020 COALA project to develop a cognitive intelligent assistant for the production line. They expect to standardize activities in the production line to reduce the gap in activity performance between experienced and novice operators. In order to set up the cognitive assistant, an operator location tracking system was needed to identify issue hotspots and sequence of activities in the production line. In this project, a suitable motion capture system was explored and deployed at Diversey Enschede 5L/ 10L production line. A literature study was performed to compare the state of the art motion capture and motion analysis methods. From the literature study results, the project decided to deploy a markerless motion capture method using Zed 2 camera.
The data collection method was tested at Enschede with Zed 2 camera which has in-built object tracking algorithms. The project applied an ethical approach to operator tracking, giving due respect to operators’ privacy concerns and anonymity. The Value Sensitive Design method was applied in this project to identify the stakeholders, their values, and the project’s future speculation. The data collection, storage and upload to cloud server was conducted using indefinitely running Python codes. The tracking was anonymized by allocating random identification numbers to denote objects and thereby, no personal data that can identify the operator were being stored. The data was captured and stored in spreadsheet format and processed using Python. The project concludes with the implementation of Z-Dash, an interactive tool that visualizes the data in various meaningful representations. Z-Dash offers graphs such as the Spaghetti chart for visualizing operator location and movements, Heat map of operator location concentration and Pareto chart that visualizes time and frequency of visited stations. The tool was evaluated with participants from Diversey to estimate the usability, interactivity and effectiveness for process improvement. The project proposes this tool for identifying the sequence of operator activities during events like changeover or stoppages, identifying issue hotspots and comparing best practices for similar events.