How does CIM leverage Artificial Intelligence?

CIM’s pioneering software architects and data scientists have conceived highly advanced applications of artificial intelligence (AI), underpinned by sophisticated data science. By combining CIM's software, UX, and engineering expertise with AI, PEAK can swiftly onboard buildings within days, enhance the productivity and decision-making of operation teams and deliver significant value in a compressed time frame.

Some of PEAK's key applications of AI include:

1. The PEAK Platform’s Rules Engine, containing thousands of rules-based algorithms to monitor plant and equipment;
2. Metadata prediction for more streamlined onboarding;
3. Sophisticated anomaly detector for highly accurate site commissioning; and
4. Automated rule deployment procedures.

Rules Engine

At the heart of the PEAK Platform is our industry-leading Rules Engine, through which millions of building data points are streamed for automated fault detection and diagnosis (FDD). The Rules Engine contains a growing library of thousands of algorithms deployed across building equipment to monitor performance. Examples include overnight operation, mechanical failure, energy wastage, safety and compliance, tenant comfort, and sensor performance. These algorithms are developed by CIM’s expert mechanical, mechatronic, and electrical engineers based on real-world building operating experience to ensure early identification and detection of energy wastage and optimisation opportunities.

The pre-configured rules-based algorithms continuously monitor all building data points, offering high-value insights to address faults, inefficiencies, and opportunities to optimise.

The PEAK Rules Engine is an ‘expert system’, a specialised form of AI designed to simulate the decision-making ability of a human expert. These systems mimic the reasoning and analytical abilities of technical experts in a specific domain, in this case, mechanical, mechatronic, controls, and electrical engineering. A core benefit offered by the Rules Engine, characteristic of an expert system, is its ability to store and process the knowledge of trained specialists, making it accessible to non-technical team members. This allows for wider dissemination of expert knowledge and permits data-driven decision-making in the complex and dynamic world of commercial property operations.

Metadata prediction

Ensuring a seamless onboarding, during which PEAK is integrated with a new building, is crucial for delivering immediate and sustained value. As such, we have invested heavily in ensuring the process is as efficient, accurate, comprehensive and technology-enabled as possible. One aspect of this is our AI-powered metadata prediction, whereby the raw Building Management System (BMS) text for a given data point is run through a proprietary model to predict what its metadata label should be, namely equipment type. The model is informed by data and learnings available through our vast network of previously commissioned buildings.

This process reflects another subset of AI known as machine learning, as it uses intelligent algorithms to interpret, predict, and process data in a way that mimics human cognitive functions. By leveraging machine learning and predictive analytics, the model is trained to analyse data and make predictions about an unknown, in this case, the appropriate metadata label for an item of plant or equipment. The model accelerates site commissioning significantly, reducing the time to first insight, expediting speed to value, and removing the inaccuracies attached to the manual efforts of competing solutions.

Anomaly detector

Another tool to enhance the onboarding process, leveraging similar principles, is anomaly detection, which identifies and corrects irregularities in data mapping. It uses semantic similarity and text-to-vector algorithms to identify if any points were missed or incorrectly mapped during the commissioning stage. This model analyses text descriptions from BMS inputs, often short, abbreviated, and lacking spaces. These descriptions are mapped to an internal schema using language modelling, converting them into numerical vectors.

The core function of the model is to assume that points with similar vector encodings, and thus clustered close together, should represent the same thing. It analyses these clusters, checks the consistency of labelling across them, and identifies outliers. For instance, if 90% of a cluster is labelled as "supply air temperature," the remaining 10% of unlabeled or differently labelled points might be considered anomalies. Overall, this anomaly detector plays a crucial role in ensuring accurate labelling and mapping of BMS points post-commissioning, enhancing the overall efficiency and reliability of the system.

Accelerated rule deployment

The effective deployment of rules is another vital component in the successful commissioning of a site. CIM’s technology team has designed a process that leverages advanced data science and analytics to automate and standardise the application of relevant rules at each property. A proprietary script developed by CIM’s software team has replaced a traditionally manual task, reducing the active work time required to deploy rules by a remarkable 90%, while enhancing accuracy. It is underpinned by a robust understanding of what points are to be commissioned per property and in what order, so as to ensure faults are identified without delay.

For each property, the technology can quickly establish what points are required to achieve at least 80% rule coverage. Instead of commissioning all points from the full BACnet scan, which can number in the hundreds of thousands, we can be far more targeted and sequential by prioritising those that will allow for the greatest number of critical rules to be deployed. The result is a more streamlined, targeted, and efficient rule deployment system, saving time and offering greater coverage, consistency, and fewer mistakes. An innovative example of technology augmenting human effort, what previously took weeks to accomplish can now be done within a few hours, further shortening the time to first insights.

How does our use of AI help to solve the problems of our clients?

• Accessibility for non-technical team members - CIM’s application of AI and ML transforms complex fault detection and diagnosis into a more accessible and intuitive process. This makes sophisticated technical information accessible to non-technical team members, effectively upskilling them and reducing the reliance on hiring of highly technical staff.
• Faster ROI - Using AI greatly accelerates the commissioning process of sites. The rapid commissioning ensures that the value from the platform is realised quickly, enabling clients to benefit from PEAK’s intelligent monitoring capabilities without the typical delays associated with traditional commissioning processes. By reducing the time to operational efficiency, these technologies directly contribute to faster ROI and immediate improvements in building performance.
• Improved accuracy and coverage - The integration of AI into CIM's processes significantly enhances accuracy and expands the coverage of automated FDD. This means that more faults across a broader range of operational plant and equipment are identified with precision. By capturing a wider spectrum of potential issues, these technologies ensure a more thorough and reliable monitoring of operations, improving efficiency and reducing equipment downtime.

Looking ahead

The collaboration between CIM’s product, integration, technology, and customer success teams drives our continuous innovation. This dynamic interplay of expertise and perspectives ensures that we are always at the forefront of incorporating even more advanced AI applications into our platform and processes. By constantly iterating and refining our approach, we are committed to exploring new frontiers in AI, aiming to deliver an even more powerful, efficient, and user-friendly experience for our clients.