Written by Fredrik Löfberg
At Quentic we are always exploring new technologies in order to understand how these could be utilized to improve workplace safety and make it more automated. Automation is key to improving safety, as technology integrations make cumbersome reporting and monitoring a thing of the past, allowing for richer data and better insights. We feel that we’ve just started the journey with our mobile-friendly approach to engaging everyone with safety. Something we’ve been looking into a lot recently is how to get more information on emerging risks on construction sites.
By no means exhaustive, the list below outlines automated technologies available today:
Visual feeds can be interpreted in real-time. This means we can identify moods and personal protective equipment from a camera at a gate.
Environmental conditions can be monitored and compared to acceptable values. This could be things like heath, humidity, and wind.
Measuring the condition of power tools and heavy onsite equipment.
Measuring the structural conditions of scaffolding and other supportive structures.
Measuring surface conditions with acceleration sensors in safety boots. With a sensor, we could recognise all cases where people did slip or stumble. It could also be used to identify who is climbing and where people are at a given time.
Combining information flows into actionable data
As the above list shows that there is a vast amount of information that can be collected. Most of this is automatically recorded, and doesn’t need to be manually tested by people anymore. As with the collection of data, the time spent collating and turning it into understandable and actionable insights is a major undertaking. Data is useless without the ability to clearly visualise statistics and trends, sorting multiple streams of information coming in at once. As the volume of available information increases, the analytics need to keep up in real time. Simply put, better, more accurate, and higher volumes of safety information brought about by automated systems will become too much for any safety manager to manually keep track of, let alone recognise the patterns and emerging risks shown in the data.
Introducing Machine Learning and Artificial Intelligence
In recent years Machine Learning (ML) has become a commodity. For example, Apple has announced it is implementing an ML chip in its new product line. In the context of construction sites, ML could take all these information flows as inputs, learn from them, and pre-process them into highly accurate predictions and preventative suggestions for humans. It can also be taken a step further and let the system make the best decisions to improve on-site HSE. Systems could communicate directly with on-site personnel if it sees unsafe behavior or wants to warn of an emerging risk. This communication could be undertaken by chat bots, which we’ve previously blogged about "Chatbots take over daily EHS Management", and are developing in the near future to make worksites in every industry safer than ever.
In the future, we will see this trend continue where the more routine and manual aspects of safety and site manager’s current tasks are taken care of by Artificial Intelligence (AI). It should be seen as a positive development, as currently many safety and site managers are so occupied with operative tasks that they don’t have time to communicate with everyone. The use of ML and AI to free up time taken by manually collecting and managing safety data means that these job roles can shift to focusing on more strategic and innovative ways to improve safety for everyone in the workplace.