The application of artificial intelligence (AI) to environmental health and safety (EHS) management is growing across virtually every industry. Once considered the stuff of science fiction, AI is becoming essential to EHS decision-makers who need tools to help make sense of the vast volumes of data generated by mobile devices, apps, computer-controlled equipment, wearables, Internet of Things (IoT) sensors, and other technologies that are increasingly relied on in the workplace. One result of this data will be a steady stream of new AI implementation projects in the coming years that will further integrate advanced data analysis and prediction techniques into our profession.

Data analysis represents a fundamental change in the focus and goals of EHS professionals. Previously, many corporate leaders aimed their attention heavily toward regulatory compliance. In recent years, that attention has shifted toward a more risk-based, proactive approach—so it is fitting that emerging data analysis technologies, including AI, are empowering EHS professionals to not only make this shift but identify previously unseen opportunities and raise the bar on EHS performance.

MEET YOUR TEAM

The most critical factor affecting the level of success of a project is communication. If you’re considering implementing an AI project, it’s critical to understand what roles are involved and how best to work with your team to ensure successful development and implementation. A typical AI project team will consist of the following roles:

Subject matter experts. Generally, SMEs are the users of the AI technology (for example, EHS professionals) who benefit most directly from the project and possess the deepest understanding of its value. SMEs consult with the team to provide the actual requirements for the project and present the basic business issue. Depending on the scope of your project, SMEs may also include other EHS professionals.

Data scientists work directly with the SMEs to develop and apply the analytical techniques and data modelling that will address a given business issue. Data scientists define the best AI models to utilize in the project and ensure that its overall analytical objectives are met.

Data engineers possess the deep technical skills to build the integration points to secure, parse, and format the necessary data and provide support for data intake into the analytical model. The data engineer works jointly with the data scientist to help build data in the correct ways for optimum analysis.

Data officers ensure that data is reliable and managed responsibly, the project’s key milestones are met on time, and the results are of the expected quality.

COMMUNICATING WITH STAKEHOLDERS

During the project approval process, it is critical to get support from key stakeholders throughout the organization, including management, which authorizes budget and resource allocation, and the workers likely to be impacted when the AI is implemented.

When securing and maintaining support from management, it’s important to clearly quantify and communicate the business value of the project. Management will be focused on the financial costs and savings that AI can deliver, so be prepared to provide some specific data on its benefits.

It is equally important to secure and maintain support from workers, even those who are not interfacing with the AI or your other EHS data management systems on a regular basis. A common fear among workers is that AI will replace them. In most cases, AI projects augment and improve processes, freeing up resources from tactical and administrative work so they can be applied to higher-valued tasks. Letting workers know about their job security goes a long way toward winning their support.

Similarly, if the AI incorporates video data inputs to synthesize insights into EHS performance, workers may feel that management is surveilling their work efforts. No one wants to be filmed when doing their job, and it’s easy to see how workers might feel uncomfortable.

For both management and front-line workers, securing and maintaining support for your AI project implementation depends on clear communication of its value. That value will be different for different stakeholders. You and your project team should be able to illustrate this value no matter the audience and serve as champions for the project’s success.