Analysis of Workload by Instructions and Fixed-Point Observation Photographs

• Published in: 2022 Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems (SCIS&ISIS) pp. 1 - 8
• Main Authors: Niimi, Ayahiko, Ito, Kei, Nambu, Misako, Harada, Yasushi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 29.11.2022
• Subjects: Cameras; Companies; Data mining; Data visualization; Employment; extract data by hand; instructions; long short-term memory; Photography; time series data; time-lapse video; Writing
• DOI: 10.1109/SCISISIS55246.2022.10001906

There is a way to proceed with work by writing work instructions on documents and exchanging the documents between departments that work in the workplace. It is possible to examine the visualization of work volume and efficiency analyzing the instruction flow. Therefore, we will visualize a part of the work and improve efficiency following the flow of documents called "instructions" used by the company. Additionally, we will develop a system to understand the work. Fixed-point photography is also performed by installing a web camera in the place where the instructions are placed in the company and taking a picture when something moves. This fixed-point photographed data is called photographic data. In this study, we propose a method to analyze the workload through the following four analyzes. 1) Analysis of workload based on the number of instructions and photographs; 2) prediction using long short-term memory; 3) extract and analyze necessary information from scan data; 4) analysis by time-lapse video. In each analysis, we analyzed the rough work; however, we could not analyze the detailed work contents.