Matthew Graham

A new approach to building maintenance, based on strategies used in the manufacturing and aviation sectors, is extending plant lifecycles while improving energy efficiency and occupant comfort, all at reduced cost to the landlord. And it’s expected to become business as usual by 2020-2025.

Developed by Matthew Graham, senior facilities manager at Knight Frank, the first implementation of “predictive maintenance” has been completed at LaSalle Investment Management’s 222 Exhibition Street building in Melbourne.

Utilising insights generated through the Bueno data analytics platform, 222 Exhibition Street’s property and facilities management team, in partnership with [HVAC maintenance contractor] Airmaster, have shifted from a reactive approach to a proactive one, where issues are identified before they turn into major problems.

This has seen a 68 per cent reduction in temperature and noise complaints since data-driven maintenance has been implemented.

An example of how the approach works is the use of vibration analysis for the HVAC, which enabled the facility managers to pick up on an issue that otherwise would have caused “catastrophic failure” of the HVAC in one instance, Graham says.

Data analysis showed an air handling unit’s motor bearings were failing, which were then replaced immediately for $3000 without inconveniencing occupants.

If the bearings had actually reached the point of failure, it would have meant spending up to $20,000 to replace a motor, on top of causing significant discomfort to occupants as it was a critical part involved in supplying 25 per cent of the entire building’s ventilation, and would have necessitated a one-week period of downtime.

As it was, the bearings were able to be replaced quickly on a Saturday, and tenants were not disrupted. In addition, root cause analysis was completed on the failed bearings and showed what was causing premature failure. As a result, earthing rings were adjusted to reduce the risk of it happening again.

No more double-call outs by streamlining maintenance and repair activities

With big data analytics and condition monitoring driving maintenance activities, technicians can use an allocation of time built into the maintenance contract to attend only assets and associated components that show signs of performing below peak performance.

The technician can then make an accurate and efficient diagnosis of the fault using current and historic data, making the necessary repairs without the need for a second visit or quote to go through an approval process.

The comprehensive analytics also means that when a HVAC technician comes to site, they can be given precise information regarding the problem.

Graham says there has been an estimated saving of around $20,000 for the calendar year, just on this aspect of building maintenance.

That then allows more hours in the budget for completing actual repairs, Graham says. As well, the technicians are experiencing a “better quality job”.

The whole goal was to do maintenance “more innovatively and effectively,” Graham says.

“I hated the clipboard process of maintenance,” he says.

By reducing the time and spend on working out what’s ineffective, a lot of the “fat” can be taken out of a standard HVAC contract and more budget allocated to minor repairs and vibrational analysis. Graham also ensured that any maintenance and testing in regards to compliance was not removed from the maintenance contract.

He says the predictive approach is standard in manufacturing, where he previously worked. This is because the cost of a production line going down is not just fixing the failed part, it is also the lost production time, which can equate to millions of dollars.

The tenants are happier

Graham says there have been fewer complaints from tenants, and any potential issues for tenants are intercepted and resolved proactively.

Comparisons of call rates to the facility response centre since the predictive regime was put in place to the same period the year before shows that temperature and airflow-related complaints have fallen drastically – from 54 work orders to 18.

Complaints about HVAC-related noise have fallen a similar amount, from 17 to five.

In monetary terms, the cost of responding to complaints has fallen from $22,000 to $5500.

The plant is also running more efficiently, resulting in energy savings of around 260,000 kilowatt-hours, or approximately 11.5 per cent compared to the previous year’s consumption.

Those savings are now being invested in the building via repairs and upgrades.

“All of these things make predictive maintenance so exciting,” Graham says.

“And because of the forward thinking approach of the landlord, we were provided with the tools to undertake this new and innovative approach.”

Gareth Sneade from LaSalle Investment Management, which owns the building, says, “The work that Knight Frank has done with predictive maintenance has led to significant and quantifiable improvements to the service levels we are now able to offer our tenants.

“Reduced downtime on services, together with lower operational costs, means a great deal to our tenants and often put the building at a competitive advantage within the market place.”

Graham is hoping to see the approach extend to other buildings.

“We are working with Airmaster to make it flexible and friendly for other sites.”

Eventually, Graham is hoping to make predictive maintenance an option across all Knight Frank-managed buildings.

An exciting industry shift

Airmaster account manager Sean O’Shaughnessy says the shift from preventative to predictive maintenance is exciting, and he expects the majority of service contracts to be “data driven and predictive” between 2020-2025.

“It’s been exciting to see the introduction of models from aviation and manufacturing,” he says.

“Planned preventative maintenance has been the norm [in buildings] for 20 to 30 years but it’s no longer the most efficient model when it comes to managing a building.”

It has also been exciting to see that the theories identified in scoping the approach have been realised in the outcomes.

O’Shaughnessy says the results have been refreshing for both building management and for technicians.

“The gains for us are gains for the client.”

One of the gains for his team has been the ability to streamline activities due to reduced workload in terms of ad-hoc callouts.

He says if the approach can be applied more broadly, it will allow the company to manage its technician labour loading more effectively across seasonal activity peaks and lows.

“It also means we now have a model that is tried and true. We can offer it to other clients and it can become a point of difference for us.”

O’Shaughnessy says that within six months of it operating, the 222X project realised all of its expected outcomes.

From a labour point of view, integrating vibration monitoring means technicians are attending assets at the most cost-effective point in time, removing the calendar-based approach.

Other clients are already looking at the model and potentially following 222 Exhibition’s lead, he says.

“The more reliable data you have, the better the solution, and the better the outcome.

“Data-driven maintenance is the way of the future – it results in energy savings, reduces callouts and results in cost-avoidance of more expensive repairs.”

2 replies on “Why predictive maintenance is a game changer for commercial properties”

  1. No question that this is a valuable concept, with a potential to improve efficiency and performance in buildings. But what is less clear is how “predictive maintenance” is new or different from practices such as continuous commissioning/optimization or the use of automated fault detection/diagnostic tools (both of which have been available for at least a decade). Can anyone clarify?

    1. Clarify “predictive maintenance” versus “preventative maintenance” versus “Re-commissioning” and “optimization”. Lets add planned and scheduled maintenance to the mix. Different companies, especially global, have different terms for simlar activity, but you know this.
      We could argue that Optimization and Energy performance KPIs or maintenance tasks should be included in a service agreement and not a seperate activity with a new title, and you know this.
      BMS companies and others are advocating building analytics, apart of “big data”, which will only ever be as good as the tech whom sets it up. This “setup” then affects the data, which then affects the predicitive maintenance and fault reporting. If the set up is flawed the data is compromised.
      But, the more we practice this and the better we get at it, the better or more accurate predicitive maintenance becomes. inturn, optimization becomes a natural & intuative task with the additonal data/information collected. And recommissioning can be better automated and/or monitored for the long term.
      My question, is who is looking at the data? who is taking the time in theie busy day to examine and refine this data to optimise the building’s systems and energy efficieny. And when is collecting and reviewing data economically viable and when does it reach critical mass that the building no longer reaps the benfits of the information versus the costs of continuelly collecting the data and reviewing it.
      Way too many words for a saturday, but id like to know if someone can clarfy the practicalites, humans still need to be invloved to make it work, and we need them to get the actual maintenance done and input that data too.

Comments are closed.