Is Intelligent Maintenance another analytics package?
We are often asked if Intelligent Maintenance (IM) is just another analytics package. The answer is no. IM is not merely an analytics package; it leverages building data and analytics rule sets as inputs to create a comprehensive end-to-end workflow. This workflow includes a virtual task-augmented technical management system enhanced by machine learning. Additionally, it features a prompt, prioritised, and proactive workflow system, which integrates with our technicians' field reporting system and utilises their feedback to guarantee that IM learns what's important for you and your building.
IM is designed to be analytics-agnostic, meaning it can work with data and analytics services from any source. Our IM engine processes the data source to drive the integrated workflow system, and we offer multiple integration methods to do so.
Let's first discuss building data and analytics for the built environment. What was considered cutting-edge technology a decade ago has now evolved into a mainstream, well-understood, and generally well-accepted science. This evolution supports the operation and optimisation of buildings, building systems, sub-systems, and individual assets such as chillers, boilers, building management systems (BMS), air handling units, lighting control, vertical transportation, etc.
Data analytics and data science for buildings generally consist of four main components:
Data acquisition, transportation, and storage
Data semantic tagging (making sense of the data)
Data analysis, correlation, and insights
Information platforms, including workflows, ticketing, and dashboards
Challenges with Traditional Analytics Systems
Many data analytics companies have great platforms designed to provide insights into how buildings operate. Although they all believe that their system is the best—more powerful, more insightful, capable of determining the root cause of a problem, predicting issues, and providing optimisation and waste avoidance insights—ultimately, they are all similar and somewhat commoditised.
Platforms that visualise data, cost avoidance engines, waste and emissions reporting, monetisation, and workflow management are generally the differentiation between analytics suppliers rather than the data collection and insights generated from the analytics.
Teams of analytics engineers, account managers, service technicians, engineers, and facility managers are involved in getting insights derived from these sophisticated systems to those who can act on the information, the service providers.
The process of "onboarding" a new building and new data now involves new tools that use predictive, AI-based capabilities to sort through point data to identify and tag the information with high accuracy and automation. Semantic tagging standards such as BRIC and Haystack built into these tools make switching between platforms relatively straightforward. The time, effort, and cost of onboarding have reduced significantly, further commoditising analytics.
IM puts every onboarded data point to work and avoids collecting and storing data that doesn't add value. Our approach ensures that the onboarding process is simplified and streamlined into our IM engine, driving an economical and effective onboarding process.
Limitations of Managed Services
Property and facility managers have adopted automation of work order management through direct M2M connections between analytics platforms and facility management platforms. Although this enables a direct connection to service providers to respond to a problem or insight, it is not part of an integrated workflow and works like a reactive service call.
The role of authorising expenditure, issuing work orders, and coordinating service providers is most often in the hands of the building owner or their appointed property or facility management providers. In some cases, data analytics providers manage the flow of work to the service providers; however, all of this involves added costs packaged as a 'managed service.'
Often, analytics systems have a prioritisation system that pushes the highest value insights to the top of a list. Cost management and budget control can determine what can be done and tend to be at the expense of automation. In other words, there is human intervention across the process, which can hold up the free-flowing resolution of issues.
Service providers are expected to update work order management systems to show that work has been done and the problem is resolved or requires further action. Adopting analytics-based insights adds to the service delivery cost, which means the value potential is not fully realised.
Additionally, there are costs such as onboarding, transportation, cloud storage and computing, IP, data management, attending meetings, working with service providers, resolving or escalating work orders, understanding if the analysis was accurate and identified the problem, recording the on-site actions, and determining if the correct service provider was sent, i.e., whether the technical discipline was correctly identified in the analysis—a controls issue that turned out to be a mechanical issue such as a broken damper linkage.
The disconnect between planned maintenance tasks included by service providers and third-party analytics insights only adds to service delivery costs, primarily driven by overlap. Technicians, account managers, and facility managers are expected to update multiple work order systems, consuming valuable time that could be spent resolving more critical issues to improve building performance and customer satisfaction.
In this typical analytics model, additional costs are offset by downstream efficiencies such as energy use and emissions reductions, tenant satisfaction through proactive service response, and long-term reductions in service call numbers. However, this tends to rely heavily on the quality of the maintenance provision and service response.
Our first foray into Data-Driven Maintenance in 2017 confirmed all the above. The result was a system that added significant costs with downstream benefits disconnected from service delivery teams. There had to be a better way, and we knew the answer lay in integrating building data analytics directly into the maintenance and service delivery workflows and switching from static to dynamic tasking and service response.
Airmaster Intelligent Maintenance (IM) – Transforming Service & Maintenance Delivery
The (r)evolution of Maintenance Service – Airmaster's Intelligent Maintenance (IM) is a second-generation system that delivers efficient, data-integrated maintenance and service delivery at an overall lower cost than existing physical service models.
The dynamic nature of the maintenance tasking shifts high-value, high-return activities to the physical technician’s service tasks, removes any tasks already completed virtually, and ensures compliance or human intervention tasks are also in the physical technician’s workflow. Task completion and reporting via the technician's field mobility system update in near real-time into a customer IM portal. A technician feedback loop in the task resolution allows for a virtuous cycle to the Machine Learning capabilities of the IM and Analytics engine.
Tasks that fall outside the customer's included priority scope of the IM agreement are automatically directed through a customer quotation system that is also fully automated using the Airmaster Copilot AI model.
In effect, Intelligent Maintenance includes virtual technicians working congruently with our on-site physical technicians to deliver more value in a "No Gaps, No Overlaps" integrated approach that is always on, always optimised, and completely transparent via the IM customer portal.
No longer is a task simply a set of predefined steps carried out by a technician who visits the site at a point in time. Tasks are now dynamically changed to reflect what the equipment, system, or building data is prioritising; task steps that are compliance-based are always included in the physical technician's workflow, and task steps that are calculable are assigned to the virtual technician's workflow and are executed continuously. Task steps that need adjustment or other human intervention, or data that indicates degrading operation or efficiency, generate new dynamic steps in the physical technician's workflow. Data that indicates an imminent pathway to failure or a safety breach triggers high-priority alerts directly to the technician's smart device.
Supporting data, correlated trends, historical service information, and other relevant information are included in the technician's appointment information to augment their capabilities and streamline their diagnostic process.
The system is a fully automated, machine-to-man-to-machine (m2M2m) platform providing a closed feedback loop that applies Machine Learning to both the root cause analytics and the workflow system. It is supported by a customer Intelligent Maintenance Portal that continuously updates and provides key metrics and reports against KPIs and benchmarks, energy, emissions, cost avoidance data, and access to detailed reporting.
By completely separating the virtual technicians' "always on" tasks and providing the data and results from the virtual technicians' activities, we can lower the time needed on-site by our physical technician teams. The virtual technician operates at a lower cost per hour than the physical technician, combining to lower overall cost while enhancing service delivery through the "always on" nature of data analytics and the IM engine.
This combination of virtual and physical services provides a lower overall Intelligent Maintenance cost, even considering the cost of analytics services, data transportation, data tagging, data storage, and analytics IP.
Not all buildings are suitable for Intelligent Maintenance; the combination of building size, asset types, and operational complexity combine to determine the cost benefits of switching to IM. However, qualified buildings produce significant maintenance cost reductions built into the maintenance contract upfront. These savings are magnified as a contract matures and amortised investment costs wash out, resulting in a cost-linear service agreement.
Intelligent Maintenance can deliver upfront maintenance contract savings of 2-20% that expand over the life of the contract. Additional savings in analytics provision costs of 30-80% can be achieved by consolidating the analytics source to Airmaster's own analytics engine, built specially for IM, due to our holistic maintenance specification approach that removes digital and physical overlaps.
The fully transparent nature of the analytics, IM service delivery, and IM Tasking consolidated into Airmaster's Customer IM Portal makes tracking the benefits and key performance indicators simple.
Conclusion and Future Insights
I hope you enjoyed reading this second instalment on analytics and Intelligent Maintenance. In my third post, I will discuss how we created IM through co-design, change management, and the automation of the data and process envelope. When changing the way our field service teams deliver maintenance and service, the importance of bringing them along for the ride and 'being in' early through design versus 'buying in' later at implementation cannot be underestimated.
For further information, please contact Eric Rodrigues, Airmaster National Business Development Manager - Digital Buildings at erodrigues@airmaster.com.au, or 0466 140 464.
About Airmaster
At Airmaster, our goal is to lead the way in integrated building services, combining HVAC&R, building automation, electrical, and fire systems to create smart buildings that are not only compliant with regulations but also optimised for energy and resource efficiency. We strive to ensure that our buildings are comfortable, functional, and tailored to support our clients in achieving their sustainability and operational goals.
Established in Melbourne in 1988, Airmaster has grown to a network of 17 branches across Australia and New Zealand. Our holistic approach to building services leverages advanced data-driven virtualisation to revolutionise how we work, delivering enhanced productivity and continuous service. This integrated model offers a more cost-effective solution than traditional methods, providing our clients with innovative, always-on support tailored to their needs.
About Noel Courtney
Noel Courtney established Airmaster 37 years ago, expanding it from its humble beginnings in Melbourne, Australia, into a nationally and internationally recognised integrated service provider. Today, Airmaster boasts 1,200 employees and 17 branches. In 2006, Noel developed the award-winning PlantPRO chiller plant optimisation system, and in 2014, he co-founded Bueno Analytics, further showcasing his innovative contributions to the industry.
Comments