Predictive analytics in construction worksites has been added to Bookmarks.
Predictive analytics in construction worksites
A GovLab report
The notion of safety is one that is easy to understand, but difficult to implement. It is simple because it is universal, and most organisations and governments alike inherently recognise this, having introduced a slew of initiatives and mechanisms to protect employees from potential hazards at work in the face of the unprecedented global attention on workplace safety and health (WSH) issues.
Yet it remains complex because resources are fundamentally limited and, as a result, choosing the right areas to target can be significant and consequential in preventing workplace injuries and, of course, deaths. Singapore, although a relatively safe place to work, still faces a number of challenges in terms of WSH issues. In the span of a year between 2012 and 2013, the number of worksite related injuries experienced a significant surge from 11,113 to 12,115. In particular, the construction sector accounted for the greatest number of fatalities, which increased from 5.9 per 100,000 employees to 7 per 100,000 employees during the same time period.
In order for Singapore to achieve a national fatality rate of 1.8 per 100,000 workers by 2018, and to hold one of the best safety records in the world – as envisioned in the WSH 2018 strategy co-drafted by the WSH Council and Ministry of Manpower to safeguard the safety and health of employees at work – more work remains to be done.
Many of the safety procedures that are in place today are a result of reactive measures taken in response to unfortunate incidents in the past. But we do not always have to learn things the hard way. In this respect, predictive analytics offers us one of the greatest opportunities. By examining historical safety incident data to identify potential trends, we are better able to take pre-emptive actions to mitigate or even, to a certain extent, prevent such occurrences.
This report presents the methodology of an iterative predictive analytics model that was developed for the Ministry of Manpower to gain insight into some of the factors that contribute to worksite accident risks for construction companies in Singapore. The model is characterised by a three-step process: gathering hindsight, deriving insights, and then, finally, producing foresight. Later, we also share some of the key features of the model and future enhancements that can be made to improve its predictive power.
While this model has been designed for a specific context, its methodology can be similarly applied to other issues regardless of industry. We hope that this publication will provide you with a glimpse into the transformative potential of predictive analytics.
Download this brochure for a summary of the iterative predictive analytics model that was developed for the Ministry of Manpower to gain insight into some of the factors that contribute to worksite accident risks for construction companies in Singapore.
For more information on this publication, please contact us here.