DESIGN & COMMUNICATION
TECH & DEVELOPMENT
BUSINESS OPERATIONS
STRATEGY, TALENT & MANAGEMENT
ENGINEERING
CASE STUDY
improving safety with computer vision
We knew we would solve the challenge, but seeing the results from the edge software and algorithms we delivered on Azure Stack Edge is fantastic. Using the existing CCTV camera networks to have one camera to track and calculate the direction and velocity of identified objects and the distance between the objects is just a fantastic accomplishment!
BACKGROUND
Ensuring worker safety in dynamic and hazardous factory and warehouse environments is crucial. Frequent interactions between personnel, machinery, and structural elements pose significant safety risks. Industry best practices, such as segregated walkways, speed limits, and strict loading protocols, help mitigate these dangers. However, human error and unforeseen events can still lead to accidents.
Recognising the need for enhanced safety measures, the client sought to leverage technological advancements for a safer and more efficient environment. The goal was to integrate lessons learned from past incidents into an open technology platform, deploying safety applications that uphold stringent data privacy and security standards.
CHALLENGE
Factory and warehouse environments present significant risks to workers, given the frequent interactions between people, machinery, and fixtures such as elevated shelving. To regulate these risks, best practices, including segregated walkways, speed limits, removal of blind spots, and prohibiting unsafe loading and unloading, are implemented. However, even with the best training and intentions, human error and unexpected events can compound, eroding safety margins and leading to accidents.
To ensure safe and efficient operations across their factories and warehouses, the client wanted to apply the lessons learned from past incidents within an open technology platform by deploying a series of safety applications while respecting data privacy and security concerns.
SOLUTION
Nexer Insight developed an intelligent factory surveillance system using Machine Learning-powered Computer Vision to automate the monitoring and protection of factory operations. This identifies vehicles and people moving within forbidden or shared zones, alerting those at risk of collision and notifying them accordingly.
By collecting examples of safe and unsafe behaviour and then running iterative ML model training, the solution can spot a wide array of potential safety violations and case-specific rule breaches, generating automatic alerts for the company’s safety system.
BENEFITS & RESULTS
With the power of Computer Vision and Machine Learning, combined with employees’ expertise, the client will significantly enhance the safety of their factories and warehouses. Thanks to the solution’s data collection and analytics capabilities, the company can target its safety awareness initiatives more effectively, using data-driven insights from actual operations. It’s a significant step forward for AI-powered safety in the industrial workplace.
More capabilities include:
− Detection of potential safety violations with higher accuracy
−The potential to reduce the number of unsafe events within days
− Boost productivity by minimising injury-related absences
To deliver, the client required a partner with deep expertise in Azure ML and Computer Vision who could design and build a scalable implementation in a high-security environment. Microsoft recommended Nexer Insight as a partner based on its proven capability to implement large-scale Azure Computer Vision solutions for global leaders in the mining, shipping, and manufacturing industries, as well as its world-leading IoT and Edge Computing skills.
Nexer Insight deployed a team to work closely with the client’s health and safety, IT, and security teams to deliver the solution to the client’s Azure tenant and edge computing appliances, thereby avoiding the movement of highly confidential data outside of corporate data security.
The solution was piloted in one of the client’s factories and deployed globally across other sites.
LOOKING FOR OUR CASE STUDY IN PDF FORMAT?
DOWNLOAD NOWCONTACT US
No pressure, just a quick chat.