INTRODUCTION
Atollogy, a pioneer in the creation of market-leading computer vision solutions, developed the ability to watch multiple areas of interest with one camera in order to analyze how people interact with equipment, vehicles, and materials. Atollogy used IM Republic’s platform to centralized multiple data sets to create one of the market’s first digital twins of physical operations. The Atollogy team was merged into ThinkIQ’s operations in 2023 as ThinkIQ Vision.
Using IMR Republic’s platform, Atollogy centralized data by capturing it to the cloud, feeding it into their artificial intelligence pipeline, and translating physical operations into analytical data. This approach provides quick and easy data access, enables a 360-degree view of the business, and supports deep and holistic analysis. By centralizing data, organizations can minimize information silos, improve data quality, and better predict the impact of emerging trends or proposed changes across different departments.
Data integration and centralized data management play crucial roles in the success of computer vision and IoT applications:
Data Integration: Using the IM Republic platform, data from various sources is combined into a unified view. In the context of IoT and computer vision, this involves merging data from sensors, production capacities, navigation systems, GIS, traffic telematics, robotics, and more. Integrating diverse data allows us to gain holistic insights, a digital twin of physical operations. For example, combining sensor data with navigation data can enhance location-based services. Real-time applications, such as predictive maintenance or security monitoring, rely on integrated data to make informed decisions promptly. By integrating data, we can optimize processes, improve efficiency, and enable adaptive responses.
Centralized Data Management: Using IM Republic’s platform allows for centralized data management, the storing and managing data in a central location. Centralized storage simplifies data access and management. Ensuring Atomicity, Consistency, Isolation, and Durability (ACID) properties is essential for reliable data handling. Centralized data facilitates analytics, machine learning, and business intelligence.
ATOLLOGY’s VISION APPLICATIONS FOR MACHINING
Atollogy pioneered the use of computer vision to analyze machine shop efficiency. In this case study they deployed cameras to analyze vertical milling operations with the goal of increasing unacceptably low utilization levels.
Even the most experienced on-site supervision and human monitoring failed to detect the leading causes of low utilization. The typical response had been to buy more equipment, in this case vertical mills that cost over $300,000.
Atollogy’s analysis showed the low utilization was caused by human issues, and not by the usual candidates: machine breakdowns, damaged tooling, or part shortages.
Exhibit 1: Vertical Milling Center
Exhibit 2: Vertical Mill Workstation with Operator Image Obscured to Protect Privacy. Notice that one camera captures five areas of interest.
Exhibit 3: Composite Analysis combining multiple data elements showed overall utilization of only 51% even with very experienced operators and equipment in perfect working order.
Exhibit 4: Sample Automated Andon Light. The light status changes automatically based on commands from its associated milling machine. Each color has an assigned meaning. There may also be a flashing signal to alert a problem or safety issue.
The analysis revealed the real culprit behind low utilization had nothing to do with the equipment. It was a problem with each operator supporting multiple mills without any means of notification when the machine had completed its assigned job. The company also realized that they needed to hire a few entry level assistants/runners, so machinists could concentrate on operating their multiple machines and not running to the inspection department or to the tool crib and could focus on running their departments.
Atollogy automated the notification process by watching the andon lights that indicated each machine status: idle, running, error, and job completed. The notification system prevented machines sitting idle while operators were working on other tasks. As a result, utilization levels increased by 15%. Most importantly, the increased utilization saves the cost of buying new vertical milling machines, or $300,000 on average.
ATOLLOGY’s VISION APPLICATIONS FOR YARD MANAGEMENT
Exhibit 5: Multiple cameras to capture truck cycle times and load weights.
Exhibit 6: Executive Dashboard from integrating data sets from multiple facilities
Computer vision collects real-time container location data in yards. This continuous stream of information helps track all types of vehicle activity. Computer vision identifies unique containers, chassis IDs, and logos on shipping containers. It recognizes visual information, such as container and chassis numbers, enabling custom logic like OCR (Optical Character Recognition). Combining computer vision with other sensors provides a fuller picture of physical activities in yards, depots, and distribution centers. Managers and operators gain better insights for efficient yard management. In summary, computer vision enhances truck traffic analysis, streamlining operations and optimizing resource utilization.
Author:
Anthony G. Tarantino, PhD
Six Sigma Master Black Belt, CPM (ISM), CPIM (APICS)
Adjunct Professor, Santa Clara University, Leavey Graduate Business School
Senior Advisor to IM Republic https://imrepublic.com
Author of Wiley’s Smart Manufacturing, the Lean Six Sigma Way
Mobile: 562-818-3275
References:
(1) Data Management in IoT: From Sensor Data to Intelligent Applications. https://link.springer.com/content/pdf/10.1007/978-3-031-24963-1_32.pdf.
(2) IoT Database Management and Analytics | SpringerLink. https://link.springer.com/chapter/10.1007/978-3-030-79272-5_13