English (United States)
Select a Language

    Asia Pacific

  1. English (AU)


  1. English (US)


  1. English (UK)

When the convergence of data, machine learning and artificial intelligence saves lives

Companies often use analytics for marketing and sales purposes. These same principles can be used in the energy and resources sector to help prevent accidents, injury or death. Read more to find out how.

Construction worker working on a construction site

Trends indicate that significant resources will continue to be spent on Health, Safety and Environment (HSE) programs in 2018,1 seeing companies naturally looking for new ways to monitor and proactively manage risk. Predictive analytics will become a key player in making better HSE decisions, preventing incidents and injuries based on an accurate picture of likely events, actions and outcomes. 

Advisian Digital has created a unique service offering that combines the experience and expertise of the WorleyParsons Group with leading data science company, SaltGrid. SaltGrid helps enterprises minimize health, safety and environmental incidents by leveraging predictive analytics.

What is predictive analytics?

Predictive analytics is the practice of extracting information in large data sets to predict patterns and trends to control future outcomes. The application of predictive analytics is often skewed towards marketing and sales purposes. But these same practices can be applied in the energy resources sector to help prevent accidents, injuries or deaths.

Laptop with data showing

How can predictive analytics prevent my next safety incident?

Predictive analytics automates the carry through of lessons learnt in safety between projects. In an industry where safety is paramount, predictive analytics helps asset owners shift from today’s reactive mode towards a proactive safety culture by delivering higher visibility into what is causing negative outcomes.

Laptop with construction worker

SaltGrid, machine learning and artificial intelligence

SaltGrid will predict events using proprietary algorithms that continuously improve over time as data is input into the platform. These events will identify risky scenarios and will highlight the most effective safety activities to mitigate the risk.

SaltGrid’s algorithms apply forecasting, identification and classification techniques through the deployment of stochastic processes, neural networks (CNN, RNN and Deeo learning), random forest and unsupervised clustering methods. Most of these algorithms are trained, verified and tuned using past historical data. They’ve been proved to be fast, stable and extremely reliable with an average accuracy measured at over 90%.

For example, one analysis SaltGrid undertakes is related to the measurement of preventive measures on incidents. SaltGrid ingests data from health, safety and environmental activities such as inspections, training, meetings and large-scale infosessions to determine the independent and additive impact these activities have on incidents. Is a large scale infosession more effective at reducing incidents than a meeting? Do those two activities make more of an impact when coupled with an inspection? SaltGrid uses deep learning to recognize the relationships between activities and incidents.

Laptop with SaltGrid screenshot

A different way of thinking

SaltGrid’s powerful, transformative predictive models allow for better deployment of your resources to proactively manage future risks rather than simply responding to past events.

For more information contact: Chris Aitken, Cosimo Spera, or Heather Stewart.

Industry Safe, ‘The Top 5 Safety and Technology trends to watch in 2018’, December 11. 2017


Subscribe to receive more perspectives