The challenges facing hard hat industries in maintaining worker and workplace safety is a major industry in itself. However, all workplaces can be hazardous and the recent statement from the International Labor Organisation about tackling workplace safety highlights this. Approximately 2.3 million people die in workplace related accidents every year and there is a pressing need to address these safety issues, worker behaviour and reporting in order to turn the tide on these preventable deaths.
A recent study from Carnegie Mellon University found that future workplace injuries can be predicted with accuracy rates as high as 80-97%. By applying powerful modelling tools based on large data sets construction companies particularly could predict and prevent workplace accidents, resulting in monetary savings, stronger safety culture and workplace productivity. How could this work and why has the Health and Safety industry not found better ways to predict accidents up until now?
Predictive analytics relies on machine learning using advanced techniques such as support vector machines, neural networks and decision trees to unlock trends and patterns. These tools are not accessible to anyone but the largest contractors as is the other element required, that is, extremely large data sets. In fact, a program of predicting workplace accidents should really be conducted by something like an industry or government body. Construction companies specifically are collecting more and more workplace safety data every day in the form of safety analyses, inspections, audits and so on. What needs to happen in order to extract usable data is for these construction companies to pool data at the national or international level. Only then will the two elements of powerful modelling tools and big data combine correctly. However, there is a danger at this level that the returning data will be so general as to be unusable at regional or site specific levels.
So what are the alternatives to predicting the future in the meantime? The construction industry and the health and safety industry have built excellent best practice models into their workplaces over the past number of decades and that work continues today with the aid of modern technology. In a sense the old adage, “Forewarned is forearmed”, is the motto of these industries and can be taken together with technological advances to help to protect workers. Anybody working in any type of hard hat industry has certainly had to attend health and safety briefings along with, more than likely, site specific training. These essential elements in highlighting hazards, staff training in hazard prevention and protocols for combating workplace dangers do have a huge impact on workplace accidents. However, they are time consuming to implement and can be made to be more consistent with online rather than face-to-face versions of same.
Tackling the ‘Fatal Four’, that account for 57% of construction sites deaths is an excellent starting place for these online orientations and online orientations. They are:
- Falls 36%
- Caught between objects 2%
- Electrocutions 9%
- Struck by objects 10%
These are relatively straightforward to highlight and institute preventative practices for and they will reap rewards in terms of saving lives which is, ultimately, the goal of all these programs. In the US in 2012, 435 construction workers lost their lives in work related accidents and it should be within our powers with the technology of predictive forecasting and online training to reduce this number significantly in the coming years.
