%20(1).svg)
When an organization is planning occupational health and safety (OHS) activities for the year ahead, budget is an inevitable part of the discussion. The head of OHS would like an extra team member, or new tools, or new software to manage OHS information. The board see such expenditure as a cost, rather than an investment. Many boards do understand the marketing value of having an accreditation, such as ISO 45001 (the international standard for OHS management) or industry schemes such as SSiP or Safe-T-Cert. Some clients require an accreditation, and hence the board can see a return on their investment.
What the board members might not appreciate is that giving the OHS team the extra resource might be the ticket for getting accreditation. As an example, we’ll consider how AI-driven computer vision can support accreditation under the four headings which form the structure of the ISO management standards (including ISO 45001) and other accreditation schemes: Plan - Do - Check – Act (PDCA).
Plan
As part of the planning process, clause 6.2.1 of ISO 45001 explains the need to establish OHS objectives that are “measurable (if practicable) or capable of performance evaluation.”
It is practicable to measure reported accidents and hence traditionally this is the most common measure of performance evaluation in organizations. Until recently, proactive measures, such as observing the activities that might lead to accidents or injuries, have been more time-consuming, and hence less practicable. For example, it is not practicable to have someone stand by a production line and count the number of times that other workers reach beyond a comfortable stretch, risking a musculo-skeletal injury. But with an appropriately positioned CCTV camera, computer vision can provide constant monitoring and feedback. With AI it becomes practicable to set new OHS objectives, based on a preventative approach.
Do
Section 8 of ISO 45001 describes the need to implement and maintain processes, including (in 8.1.1) the requirement that organizations maintain ‘documented information to the extent necessary to have confidence that the processes have been carried out as planned.’ The standard explains in a note that ‘documented information can be in any format and media’ and this includes text, images and video. Traditional approaches to this involve completing checklists for inspections, or making notes of observations. Handwritten notes require transcribing or scanning if they are to be retrieved and referred to later, and can result in a pile of untraceable checklists.
Computer vision makes it practical to collect and store short, time-stamped, geo-located video clips of processes, and use these as part of the ‘documented information’ within your assurance process.
Check
Many organizations ask workers to report near misses, and see this as a key part of the ‘check’ process. However, it’s clear from Section 9 of ISO 45001, Performance evaluation, that a near miss reporting system isn’t enough. For accreditation the organization needs practical ways of monitoring activities where hazards have been identified, and of checking the effectiveness of controls. Computer vision provides a useful tool for this. For example, if you have a zoning system in place to separate vehicle routes from pedestrian routes, people might comply while you are watching, but how will you check it is working the rest of the time? Rather than waiting until a driver reports a near miss when a pedestrian gets too close to the vehicle – or worse, waiting for an accident – computer vision can let you know every time a vehicle or pedestrian strays into the wrong zone, even when no one else is there.
Act
Where ‘Act’ differs from ‘Do’ is in making use of feedback to improve the way in which OHS is managed in the future. Section 10 of ISO 45001 explains the need to be able to demonstrate continual improvement.
Computer vision and other forms of AI can collect a lot of information, which could be overwhelming if not managed appropriately. You need a system that compiles your information in one place and allows you to customize how it is sorted, filtered and categorized. You want to make comparisons over time and between locations. For example, were problem postures identified less often on production line A where new handling equipment had been provided, than on line B, using the old equipment?
Being able to auto generate safety reports full of evidence about what is working - and what still needs work - will provide compelling evidence to an accreditation auditor that your organization is improving. Automated workflows make sure that the right people – including board members - get the information they need without you spending a morning on emails. If the board members are interested in getting accreditation, they will soon see how AI will help them with that goal.
To learn how Protex AI's computer vision and AI technology can support your organization's occupational health and safety objectives and help you achieve accreditation under ISO 45001 and other industry schemes. Watch our demo video today!
.webp)
.webp)
