Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Capacity Constraint Analysis Using Object Detection for Smart Manufacturing

Hafiz Mughees Ahmad, Afshin Rahimi, Khizer Hayat

TL;DR

The paper addresses capacity constraints in manufacturing driven by labor shortages post-COVID by proposing a non-invasive framework that combines CNN-based object detection (YOLOv8) to identify chairs and workers with DeepSORT-based tracking to build object lifecycles on the shop floor. The methodology trains YOLOv8 on multi-station data and deploys it on-site to perform capacity analysis, including a cycle-time pipeline and real-time productivity metrics. Key findings include Station C achieving 70.6% productivity over six months and detailed hourly, daily, and monthly insights that reveal persistent labor shortages and bottlenecks. This work demonstrates a practical, data-driven approach to monitor and optimize capacity in manual manufacturing, enabling better labor and inventory planning and paving the way for predictive and prescriptive extensions.

Abstract

The increasing popularity of Deep Learning (DL) based Object Detection (OD) methods and their real-world applications have opened new venues in smart manufacturing. Traditional industries struck by capacity constraints after Coronavirus Disease (COVID-19) require non-invasive methods for in-depth operations' analysis to optimize and increase their revenue. In this study, we have initially developed a Convolutional Neural Network (CNN) based OD model to tackle this issue. This model is trained to accurately identify the presence of chairs and individuals on the production floor. The identified objects are then passed to the CNN based tracker, which tracks them throughout their life cycle in the workstation. The extracted meta-data is further processed through a novel framework for the capacity constraint analysis. We identified that the Station C is only 70.6% productive through 6 months. Additionally, the time spent at each station is recorded and aggregated for each object. This data proves helpful in conducting annual audits and effectively managing labor and material over time.

Capacity Constraint Analysis Using Object Detection for Smart Manufacturing

TL;DR

The paper addresses capacity constraints in manufacturing driven by labor shortages post-COVID by proposing a non-invasive framework that combines CNN-based object detection (YOLOv8) to identify chairs and workers with DeepSORT-based tracking to build object lifecycles on the shop floor. The methodology trains YOLOv8 on multi-station data and deploys it on-site to perform capacity analysis, including a cycle-time pipeline and real-time productivity metrics. Key findings include Station C achieving 70.6% productivity over six months and detailed hourly, daily, and monthly insights that reveal persistent labor shortages and bottlenecks. This work demonstrates a practical, data-driven approach to monitor and optimize capacity in manual manufacturing, enabling better labor and inventory planning and paving the way for predictive and prescriptive extensions.

Abstract

The increasing popularity of Deep Learning (DL) based Object Detection (OD) methods and their real-world applications have opened new venues in smart manufacturing. Traditional industries struck by capacity constraints after Coronavirus Disease (COVID-19) require non-invasive methods for in-depth operations' analysis to optimize and increase their revenue. In this study, we have initially developed a Convolutional Neural Network (CNN) based OD model to tackle this issue. This model is trained to accurately identify the presence of chairs and individuals on the production floor. The identified objects are then passed to the CNN based tracker, which tracks them throughout their life cycle in the workstation. The extracted meta-data is further processed through a novel framework for the capacity constraint analysis. We identified that the Station C is only 70.6% productive through 6 months. Additionally, the time spent at each station is recorded and aggregated for each object. This data proves helpful in conducting annual audits and effectively managing labor and material over time.
Paper Structure (15 sections, 2 equations, 6 figures, 3 tables)

This paper contains 15 sections, 2 equations, 6 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. You Only Look Once
  4. Object Detection in Smart Manufacturing
  5. Capacity Constraint Analysis
  6. Cycle-time Study
  7. Methodology
  8. Dataset Description
  9. Training
  10. Evaluation Metrics
  11. Post-Processing
  12. Results
  13. Insights Into Manufacturing Facility
  14. Challenges
  15. Conclusion

Figures (6)

  • Figure 1: Complete yolov8 jocher_yolo_2023 architecture consisted of backbone and head. C represents the convolutional block, U is the upsampling block, and C2F is the cspnet with two convolutional layers. A detailed diagram can be found in king_brief_2023.
  • Figure 2: Complete pipeline of the proposed methodology.
  • Figure 3: Precision, Recall, mAP@50 and mAP50-95 of the yolov8 jocher_yolo_2023 model training.
  • Figure 4: The view of floor lines where power wheelchairs and workers are represented by yellow and red boxes, respectively.
  • Figure 5: The pie chart representing the status of Station C over 6 months.
  • ...and 1 more figures

Theorems & Definitions (4)

  • Definition 1: Station Productivity
  • Definition 2: Non-Productivity
  • Definition 3: Downtime
  • Definition 4: Idle Time