The Confluence Between Human and Machine in the Prioritization of Risk in Life Safety

Here’s how human intelligence and machine capabilities can be used in facility management.

• By Matthew Icenroad
• Jun 22, 2023

In an increasingly interconnected world, the convergence of human intelligence and machine capabilities has revolutionized numerous aspects of our lives. The recent growth of artificial intelligence and machine learning tools such as ChatGPT, BARD and natural language processing have taken our lives by storm, bringing forward fast analytical processing and research capabilities.

One crucial domain where the synergy of human and machine holds immense potential is in risk prioritization. By combining the cognitive abilities of humans with the computational power and analytical capabilities of machines, we can enhance our understanding of potential hazards, improve response mechanisms and ultimately save lives. 

Risk Prioritization in Facility Management

Facility management encompasses the management and maintenance of physical assets, such as buildings, equipment and utilities. Like any business activity, it involves inherent risks. Effectively managing facilities requires a comprehensive understanding of these risks, the ability to prioritize them, and the development of strategies to mitigate or manage them.

Prioritizing risks in facility management involves assessing the likelihood and potential impact of various risks and determining which risks are most critical to the organization’s operations. This requires an understanding of the facility’s physical assets, operational processes and potential threats to building safety. 

The National Fire Protection Association (NFPA) requires organizations across the United States to maintain their buildings and life safety features through regular inspections, testing, and maintenance, which helps to identify any deficiencies. Organizations will often retain a third-party fire protection engineering firm to conduct code assessments on buildings to ensure compliance and identify any deficiencies in need of correction. Depending on the building’s size and complexity, several hundred deficiencies may be identified, leaving facility directors to ask the question, “where do I start first.”

Approaches to Risk Prioritization

Risk Matrix. One approach to risk prioritization involves using a risk matrix, which categorizes risks based on the likelihood, impact, number and concentration of findings. High-likelihood and high-impact risks are considered critical and require immediate attention. But what if an organization has numerous low-impact risks in a concentrated area? Does that increase the overall level of risk and prioritization?

For example, while a single small penetration in a fire barrier may not raise the highest alarm for immediate correction, numerous penetrations in a single area may impact the fire barrier to an unrealizable level of life safety protection. This would increase the risk and therefore increase the level of prioritization for corrective action.

Risk Assessment Framework. Risk assessment frameworks can also be used for risk prioritization. This involves systematically identifying, analyzing and evaluating risks to develop effective risk management strategies. Often utilized when a smaller set of deficiencies or tasks are being evaluated, this approach provides a customized strategy to support the organization’s efforts.

This strategy has seen success when used during construction and renovation projects, as it allows project managers and fire protection professionals to actively assess risk while projects continue toward occupancy. By assessing individual risks throughout the process, engineering staff can work with team members to adjust plans and designs to mitigate risk post-construction and throughout the building’s life cycle.

Utilizing Machine Learning Capabilities to Prioritize Risk

The identification of risk is not a foreign idea. But how can machine learning capabilities support our efforts and revolutionize our work? Facility managers must identify bottlenecks, redundancies and inefficiencies to develop strategies for improvement. While Lean and Six Sigma methodologies can be employed to streamline processes, reduce waste and improve process flow, it may take time for facility managers to identify and develop these processes and skills. Machine learning capabilities can significantly contribute to facility management by aiding in risk mitigation, process optimization and predictive modeling.

Machine learning is an aid and not a full replacement for the human element. It is important to recognize that people are still critical in the decision-making process. 

In risk mitigation, machine learning algorithms can analyze data from sensors, historical documentation, and other data sources to identify potential equipment failures or maintenance needs. Proactive maintenance plans can then be developed, minimizing downtime and reducing risk. Machine learning can also analyze energy consumption and occupancy patterns to identify energy-saving opportunities and process improvement, optimizing efficiency while maintaining a safe environment.

Machine Learning Tools Assist in Decision-Making

Using machine learning tools to optimize processes in facility management can help improve efficiency, reduce cost and enhance service quality. Through the use of these tools, artificial intelligence can help us prevent interruptions by prioritizing tasks and improvements based on all essential factors, such as level of risk, the concentration of deficiencies, funding, manpower, specific skill sets and organizational activities.

There are already a number of digital tools available to help improve team efficiency and effectiveness. Some have the ability to cut through complex data to screen and develop prioritization and processes for incoming condition reports. By using explainable predictions and consistency among organizations, these tools work in collaboration with personnel in their decision-making.

It is important to recognize that machine learning is not a full replacement for the human element. People are still critical to the decision-making process. Supervisors are still required to review results, guide algorithms and override the machine learning capabilities as necessary. This not only helps the organization maintain control of its risks and prioritization but also allows for the machine to learn over time and improve itself.

The collaboration between human expertise and machine learning in facility management ultimately enhances decision-making processes, risk mitigation and operational efficiency. By combining human contextual understanding, critical thinking and intuition with machine learning's data analysis capabilities, facility managers can make informed decisions that prioritize risks, optimize processes and improve long-term planning.

Embracing the Synergy Between Humans and Machines

Prioritizing risks and optimizing processes are vital aspects of effective facility management, and machine learning can significantly contribute to these areas. By leveraging machine learning algorithms, facility managers can identify and mitigate risks, streamline processes, reduce costs, and forecast future facility needs. Embracing the synergy between human intelligence and machine capabilities will enhance facility management practices, leading to safer, more efficient and cost-effective operations.