THE BLOG

Machine Learning: The Future Of EHSQ?

29/08/2017 15:24

Machine Learning: The Future of EHSQ?

By Pam Bobbitt, Director of Product Marketing at Cority Software

Machine learning is everywhere. You may not know it, but from fashion to fintech, machine learning is the force driving innovation and powering growth around the globe. But what exactly is it? How does it work? And more importantly, what does it mean for the Environmental, Health, Safety and Quality (EHSQ) industry?

Let's start at the beginning. The term 'machine learning' is used to describe a computer that has the ability to learn without being explicitly programmed to do so. It's sometimes described as being synonymous with artificial intelligence (AI), but in reality, where AI is the broader concept of machines being able to carry out 'smart' tasks, machine learning refers to the idea that machines should be able to learn autonomously.

Machine learning involves a computer making data-driven decisions based on sample inputs and drawing new conclusions based on previous interactions. Once residing at the cutting edge of scientific thought, machine learning has slowly but surely developed in sophistication and accessibility, becoming a silent partner in our everyday lives.

Siri's suggestions for dinner? Machine learning. Your Uber's arrival time? Machine learning. That perfect song recommendation from Spotify? You guessed it. With such a diverse range of practical uses, the arrival of machine learning in the EHSQ space shouldn't come as a surprise.

The technology is set to dramatically change the way workers operate in the EHSQ industry. In industrial hygiene, for example, machine learning could offer an effective way of crunching large volumes of data to build predictive modelling solutions that could dramatically improve efficiency. For the moment though, the technology is sill teetering on the edge of delivering a truly transformational breakthrough in the sector.

Instead, it's better to consider that the work happening today is laying the foundations for the future, building a platform for an AI solution that truly applies the power of machine learning to EHSQ. In five years time, we believe we will see predictive modelling across all EHSQ practices, allowing users to draw on software powered by machine learning to fundamentality improve decision making and data analysis.

Take a typical safety manager. By utilising machine learning, he or she will be able to increase the pool of available information to inform decision-making about any given safety practice that puts personnel at risk. They will be able to use software to isolate nominated variables to determine the effect (if any) on a particular group of employees. This is something we characterise as 'augmented decision making', a process that we believe will lead to a reduction in accidents and injuries at work.

But as with any and all new technologies, we must make sure to carefully examine its likely effects on the sector at large, committing to a considered look at its impact across the industry. With this in mind, we must implement machine learning in a way that complements the role that EHSQ officers play, ensuring that any new technology supports rather than disrupts.

Today, a supervisor is likely to spend more time on manual tasks, whereas in the future, their emphasis will be on observing workers and refining policies - with computers taking on the administrative and technical heavy lifting. This sort of collaboration will define the human-AI relationship, supporting a new era of augmented decision making that will deliver relevant and refined results by combining human initiative with AI analytical capability.

Thanks to AI - the future looks bright for EHSQ. While there are a lot of exciting technologies coming over the horizon, none can match the revolutionary potential of machine learning and AI. By marrying human ingenuity with robotic technical proficiency, machine learning solutions are set to transform safety and quality control in the workplace.

Comments

CONVERSATIONS