Monitoring Environmental Compliance using Artificial Intelligence
Environment compliance today is not a matter of choice. Increasing global warming, drastic climate change, and even Covid-19: the global health pandemic has forced countries to develop stricter environmental regulations and their enforcement. Non-compliance with these rules and regulations results in hefty fines, cancellation of permits, and sometimes even jail. Based on the nature of your organization’s operations, it is essential to identify what compliance constitutes for you, what is the jurisdiction/s in which you operate, and the regulatory landscape for the services and products you provide.
Conventionally, this involves hiring an EHS professional/consultant who can map your organization’s operations to the legal landscape. The person should also review the relevant regulations and permits and provide you with detailed and specific requirements demanded by the respective compliance. These requirements are then assigned ownerships, deadlines, limits, reporting triggers, etc. The whole process can take many hundreds of manual hours to complete. And the accuracy and consistency in interpretation will depend on the professional’s skill level.
AI assisting environmental authorities
The use of technology by environmental authorities is not new. In 2004, Charles Costello from the Massachusetts Department of Environmental Protection (DEP) used satellite imagery to track and identify 3,000 locations of unpermitted wetland fillings. The process was later adopted by DEPs of other US states. Artificial intelligence, the ubiquitous tool of this century has been recently employed by researchers to automate environmental inspections, making it difficult for businesses to slip under the radar of environmental authorities.
For example, Handan-Nader, Ho, and Liu from Stanford University trained a deep learning model to recognize unregistered large-scale, concentrated animal feeding operations (CAFO) from satellite imagery. They used high resolution satellite imagery provided by National Agricultural Imagery Program, along with the CAFO locations in the area from the information collected by NGOs to train a deep learning model.
Some other worthwhile examples include: Canadian Space Agency, funds the Integrated Satellite tracking of Pollution project, with an aim to detect oil spills. They identify suspected oil spills using satellite imagery, the information is then fed to surveillance aircraft. CleanSeaNet, a European satellite-based oil spill, and vessel detection service, monitors’ potential oil spills for European nations since 2007. The technology has been used to identify not just oil spills, but habitat fragmentation, deforestation, and even air pollution monitoring.
AI for Earth and other efforts
Realizing the importance of environmental compliance, many big companies have invested heavily in using AI for good. Microsoft funds the AI for Earth program, providing grants to projects that use AI to monitor, model, and manage Earth’s resources. More than 230 grants have been awarded until now, and many projects are running around the globe. The projects involve using AI for anti-poaching activities and using IoT and AI to improve agricultural yield.
AI is also being used to manage renewable energy. From preventive maintenance of wind turbines to predicting electricity load demand. Today AI is also being used to produce more electricity by incorporating real-time weather and operational data. Wind companies are using AI to adjust propellers’ speed based on the wind speed and direction, resulting in more electricity production per rotation.
Google had made use of AI to predict when its data centers’ energy was most in demand. By analyzing the data to predict when users are more likely to use data-intensive sites like YouTube, they could optimize the cooling needs, which resulted in the reduction of energy consumption by 40%.
The smart city concept being adopted in many cities to manage traffic, parking, and city waste management has reduced air pollution and resulted in efficient energy use. In China, IBM’s Green Horizon project uses an AI system to forecast air pollution, keep track of pollution sources, and suggest potential strategies to deal with it. For example, it can tell that for reducing pollution, given a choice between restricting the numbers of drivers on the road or closing a certain power point, what would be more beneficial in reducing the pollution in a particular area.
AI has been used by Indian farmers to obtain a 30 percent higher yield of the groundnut crop. The system provided the farmers with information on the preparation of land, applying the fertilizer, and selecting the sowing days to get a higher yield per hectare of the land. Norway is using AI to create an electric grid- which is more flexible and autonomous – such that it can integrate more renewable energy into the grid system.
AI the Game-Changer
With the satellites always watching, it makes sense for companies not to risk environmental inspections. The traditional way to EHS compliance is time-consuming, error-prone, and costly. Employing enterprise software applications that can integrate AI can be a game-changer for EHS compliance. The use of AI can increase the efficiency and effectiveness of regulatory compliance programs across various industries.
The companies can use their historic data to train AI agents which can predict and prevent hazards before they occur. The use of natural language processing tools like BERT, GPT can help to automate interpretation and analysis of compliance documentation requirements. The AI can augment the EHS professionals by automating the workflow, keeping a track of compliance activities and their respective thresholds.
The road ahead
Environmental compliance is not optional, it is not only a legal requirement but today also a moral and ethical one. The advanced AI monitoring tools, further make it impossible to slip and hide. Additionally, customers, investors, and employees prefer a company that is doing its bit for the earth. Customers use their spending power to bring the change they want. Employees want to work with organizations supporting green initiatives, and investors want to invest in companies adhering to environmental regulations.
AI tools can be deployed to analyze, monitor, and prevent environmental hazards. The network of sensors, along with edge computing and AI can track and alert stakeholders before the specific regulation violation. There are a large number of companies offering AI-based EHS compliant services. Companies like Intelex, SAI360, EASE provide data-based EHS compliance. AI in monitoring environmental compliance can help in reducing human errors, compliance costs, also it can be used to track compliance audits and certifications.