Artificial Intelligence in Facility Management

FACILITIES MANAGEMENT CHALLENGE — DO MORE FOR LESS

Facility management is the part of the business that has always been under pressure to ‘do more for less’ and to deliver the magic 10% cost savings that the core business demands of it.

As businesses start the slow road to recovery and begin to emerge from the pandemic and enforced lockdowns, facility management and its associated costs will again be under the microscope.

Traditionally, these cost savings have come from market testing, outsourcing, re-tendering, re-scoping, head count reduction and other areas of efficiencies that have by now, challenged the simultaneous demand for improved service quality and performance.

Whist technology has played an important part of facility management for some time now, through a hunger for data to measure performance and through BIM and SMART or intelligent buildings, enabling informed decisions to be made, there is now a new opportunity for the use of technology in facility management and this is arguably the biggest opportunity yet.

Step forward the use of Artificial Intelligence in facility management

Facility management is involved across every organisation, and markets across both the private and public sectors and in commercial and non-commercial entities. To optimise processes and operations, it is crucial to implement workplace practices that ensure a constant flow of communication, knowledge sharing, productivity and innovation. FM people are always busy with responsibilities and tasks ranging from coordinating teams, attending meetings, answering urgent emails, consulting vendors or suppliers, maintaining security, managing assets, etc. this list is endless.

The use of Artificial Intelligence in facility management sector precisely determines the areas where maximum effectiveness can be attained even without extensive human involvement or related expenditure.

ARTIFICIAL INTELLIGENCE (AI) — MEETING THE CHALLENGES FACED BY FACILITY MANAGEMENT

Several economic factors have increased the desire to outsource more and more of the non-core business functions, generally known as facilities services. Facilities management is ripe for disruption: it lags other functions such as production equipment maintenance by both digital maturity and penetration of technology. Although technology is available for facilities management, several obstacles have inhibited adoption, such as a lack of digital skills within the function, other priorities for leadership, and a focus on continuous cost cutting. These factors have made the facilities management outsourcing market attractive for leading vendors that were already engaged directly or indirectly with this function.

Extensive expenditure, energy utility, human error, security evidence etc. can be considered as the major issues/challenges.

Considering the COVID-19 pandemic, which has already dramatically disrupted global businesses. Oil prices have fallen, the stock market has tumbled and brands that rely on Chinese manufacturing for their products foresee delays or other disruptions in the coming months if factories remain offline. The escalating coronavirus outbreak and what it means for your business, especially in the facility management sector is to be seriously pondered upon.

There will be changes in customer and third-party behaviour, as well as changes in demand for products and services. Additionally, workforce unavailability due to sickness and changes in productivity due to increasing numbers of remote workers.

It is likely that many companies will continue to operate remotely and the ‘work from home’ policy. For those that are slowly returning to their workplaces however, effective facility management will be key and will be instrumental in getting the business back to where it needs to be. Although there is a feeling that many organisations will continue the ‘work from home’ policy there is increasing acknowledgment that more and more organisations will eventually return to ‘normal’.

The Internet of Things (IoT) has had numerous applications during the COVID-19 era, including playing an important role in the establishment and support of remote working, stock management, healthcare, and industrial manufacturing. This trend is likely to continue and evolve, as key players fine-tune their IoT devices to improve efficiency and accuracy. The opportunities for adaptation and use in facility management is endless. Sensor technology and analytics can be used to monitor desk occupancy, washroom distancing, and cleaning regimes. Temperature and humidity sensors can be installed to improve comfort and minimize the chances of spreading the virus through air-borne particles.

THE ROLE OF AI IN FACILITY MANAGEMENT

1. INCREASE COST SAVINGS AND DECREASE ENERGY USAGE

AI is completely changing the way some organisations cut costs on vital areas of operations like energy, HVAC, security, and other systems through reinforcement learning or trial and error. The AI algorithm uses the information from historic data to build a database of conditions designed to mimic the human mind.

Google has been using AI to monitor one of its data centres for the last few years, and the results are significant. Since allowing the AI algorithm control operational plant conditions, they have saved circa 40% on powering the data centre’s cooling system. With AI, not only are organisations able to save on essential operations, but employees can work more efficiently.

2. FOCUS ON PREDICTIVE MAINTENANCE

Predictive maintenance differs from preventative maintenance in that it takes on a more proactive approach. For example, instead of receiving alerts when your system is due for maintenance following a timeline, predictive maintenance will alert you when something needs to be done, even if it is not the time for scheduled maintenance. With AI, facilities teams can better determine when to leverage as-needed maintenance, instead of performing regular and arguably, unnecessary maintenance.

By implementing AI to monitor their maintenance efforts, organisations like Maximo APM from IBM have saved, on average, 50% on preventative maintenance, reduced the number of hours they spend on preventative maintenance by 50–70%, and have lowered the number of breakdowns by 70%.

3. REDUCE THE POTENTIAL FOR HUMAN ERROR

Organisations can reduce the risk of human error through AI algorithms. Not only is AI good for optimising existing office space but can be used for space planning as well.

For example, with AI, FMs can get the most accurate idea of how much space they need based on the number of employees, usage data, and more metrics as sensors can be used to forecast space requirements due to the backlog of occupancy data they store and can shed light on factors like space demand and usage.

Open Sensors is a technology company that helps organisations better understand their space usage with sensors that place sensors on employee desks and in meeting rooms to gauge use. As a result, the company was able to learn about their space usage, which led to opportunities to cut costs, rework scheduling among departments, and even redesigning their workspace.

4. STREAMLINE THE INTEGRATION OF THE IT, FM, AND HR DEPARTMENTS

AI can make systems integration simpler by improving communication between these departments, cutting costs, and breaking down barriers that prevent these teams from working together. AI can be used to understand space usage, streamline communication, improve collaboration, and make better use of data can turn these traditionally isolated departments into one powerhouse team.

5. IMPROVE BUILDING SECURITY AND PROTOCOL

For cybersecurity purposes, AI helps detect and prevent online threats and requires less human involvement by flagging potential risks by learning what is “good” and “bad” behaviour online. Also, AI can present possible solutions to problems or threats, which takes much of the guesswork out for humans. AI can be used to protect sensitive data like employee and customer information, organisational data, and more.

More than 75% of facility managers who responded to a benchmarking survey said that health and safety was a major operational concern. Minimizing contact with potentially contaminated surfaces is crucial for a hygienic building environment. This trend is likely to continue. Over 68% of facility management professionals surveyed said that improving building image was a key operational issue. Facility image and reputation can directly influence tenant satisfaction and occupancy levels, so the concern is understandable. Key cards and fobs introduce significant overhead costs to facilities. Access tokens are often lost, forgotten, or stolen and can be cloned simply. It takes time to replace and distribute new cards, not to mention they are a continuous cost. Access control that uses mobile credentials and/or biometric characteristics for access effectively solves these issues. Contactless visitor management systems, which go together with access control, provide operational efficiency. By using contactless visitor management systems with AI technology can help us to verify guests through cameras. Remote access capabilities thus also promote a healthy environment for social distancing.

AI FACILITY MANAGEMENT DEVELOPMENT PROJECTS

In the second part of this article, we take a brief look at some of the specific areas that Evbex is reviewing and developing.

1. Performance analysis and Predictions:

For performances related to energy monitoring we can collect all the data and can carry out performance analysis presented in an attractive UI Dashboard and can also predict the future.

Evbex is currently developing a facility management platform called FM Navigate which is a cloud service which can collect the data periodically, do analysis and can be visualised on a live dashboard. The data collected can be used as a source to develop machine learning intelligent models to predict and forecast.

How can we predict the future values?

For each measured value, let us take Electricity as an example. There are many factors on which electricity depends, which includes the production plan, number of currents using plants, number of hours working etc. As a result, there will also be a time series change. Usually, there will be a pattern of increase or decrease or no change at all with the time. By obtaining historic values and through training a Machine Learning (ML) model, future usage can be predicted.

An ML model will be trained to prioritise importance to various features, according to its correlation with the output. Also, the prediction can be variables such as,

• How much energy will be utilised next month if this usage is continued (so that they can do necessary steps to minimize it)
• When will be the exact time to change the next filter based on the previous knowledge of ML model and current values without the need of checking regularly?
• Predict the amount of energy that is being wasted.

2. Equipment predictive analysis

Predictive analysis on how, when and what features affect the equipment lifetime and their reliability along their life cycle. This will establish insights regarding upcoming maintenance, reduce unexpected downtime and outages, add predictability to the budget and maintain occupant comfort. With exploratory data analysis we can predict the reason for early damage of assets and estimate the quality of different assets based on their manufacturers.

3. Modelling leading actions through SMART Key Performance Indicators (KPIs)

Through establishing SMART KPIs that measure every aspect of the facility services delivered by either an outsourced or an in-house FM model, AI and ML will enable us to predict future outputs and probable outcomes. Early intervention can accelerate positive outcomes and change the course of potentially negative or undesired outcomes.

Cleaning

There are several areas where innovation and value can be applied for soft services such as cleaning.

a) People counting:

Using surveillance camera-based analytics along with the power of computer vision and deep learning, we can count the number of people across a workstation. This can provide us with many insights which will be extremely useful for proactive cleaning regimes. This includes,

1. Detection of areas with more crowd which needs to be cleaned regularly.
2. Identifying optimum time in a busy place to clean without affecting the normal functioning.
3. Create a proactive and optimum cleaning schedule that is agile and flexes to meet the occupation and traffic through the buildings.

b) Waste Segregation:

Computer vision can be used to automatically segregate different types of waste. This can make waste management an easy and much hygienic process. By this method, greater productivity can be achieved without continuous monitoring.

4) Face recognition for attendance and access

Instead of using card-swiping or manual attendance marking, the use of cameras, (contactless approach) incorporated with face recognition and deep learning can cut labour and automatically register the presence or absence of an individual.

5) Security

There are a range of AI applications related to safety and security, though many are nascent. Instead of using a key card for access control, using facial recognition we can make computer vision and deep learning to do the work with surveillance cameras. When it comes to keeping an organisation physically safe, AI can be used to boost an existing security system.

Deloitte’s Robocop, located in the Edge smart building in Amsterdam, can detect threats, notify security guards of actual issues or false alarms, monitor areas of the office, and more. It can either be controlled by a remote or automatically.

Apart from physical security of organisations, the additional sectors where AI can be inculcated in security systems could be:

1. Anomaly detection: can automatically identify and alert in case of any emergencies like fight, explosion, robbery etc.
2. Tracking technology for various objects such as dangerous weapons and sometimes face recognition across all video frames.
3. Vehicle detection and number plate identification-based entrance system. This is of course already widely used.
4. Suspicious package and similar object detection.

As also explained earlier in the blog, cyber security involves the applications of AI that detects threats and hence provides a safer virtual environment for users.

6) Covid-19:

1. a) Social distancing:

This AI technology tracks employees to enforce social distancing. The Social Distancing Alert System uses existing IP cameras and CCTV cameras combined with Computer Vision to detect if people are at a specific distance and adhere to social distancing or not.

1. b) Mask Detection:

This system enables one to identify who is without a required face mask. These systems work with existing surveillance systems along with innovative neural network algorithms to check whether a person has worn a face mask or not.

7) Other potential uses in FM:

For HR and recruitment, sometimes we often need to filter specific requirements from many applications. What do we have? We know the skill set we require, or we know the education requirement that we want. How can AI help us? AI can pair resumes automatically using NLP techniques and extract all the relevant entities. This information can be used for scoring automatically and shortlist the applicants. This can save a lot of time and improve productivity.

CONCLUSION

AI and ML technology applications for facility management services is endless and can sometimes be overwhelming but before long, every organisation will embrace and reap the benefit that AI and ML can bring in whatever capacity that is required. Organisations will inevitably start the AI journey with small steps.

AI will play such an important role in facility management services that it will be considered as the next Industrial Revolution.

Traditionally, facility management has struggled to gain the credibility it deserves across the business and most sectors.

As the core business emerges from the pandemic in fighting mood, it will look to facility management as it always does to initially ensure its survival and its prosperity in the challenging times ahead. Having exhausted everything at its disposal to date to reduce operating costs and optimise efficiency, AI will undoubtedly provide the facility management road map for all organisations and in doing so will elevate itself from a cost cutting commodity that will be viewed through a different lens as a value-adding, strategic entity that could well determine the success or how successful a business can be.

Facility management — your invitation to the top table awaits!

Evbex is an estates, property and facilities management consultancy, advisory, technology and training company that is headquartered in the UK and has bases in Asia, the Middle East and North America.

Our FM Technology platform, FM Navigate is an intelligent Facilities Resource Planning and Management System (FERP) developed to manage and administer complex and dynamic estate and facilities management services. It provides a multi-platform, cloud-based technology solution that caters specifically for the needs of modern-day challenges and future requirements, organisations through a need to support the core business in a cost effective, agile, and optimised way. Take control of your FM operations and outsourced services with FM Navigate.