This is the second part of a two-part series. Read part one here: Link.
The Diagnosis of Implementing AI to Facilities Management
Our first post was a diagnosis. This post is focusing more on implementing AI in facilities management in Malaysia.
We said AI is a multiplier that amplifies what is already there. Most FM operations in Malaysia are not structured enough to make that amplification useful.
Despite the status quo, we should not stop the AI conversation. It is a point where we should have this discussion more honestly.
The reality is buildings are growing in complexity. In design and in operations.
Waiting until the fundamentals are perfect before engaging with AI is not a strategy. It is avoidance masquerading as prudence.
The path forward for the FM industry in Malaysia is parallel progress. Fixing the fundamentals while adopting AI. Not one before the other.
What Parallel Progress Actually Looks Like in FM
1. Start with asset knowledge because it is the entry point for everything else. Every AI system, every predictive model, every condition monitoring tool needs to know what it is managing. Building that register is not going back to basics. It is building the data layer that makes every future tool smarter.
2. Record operational tasks when it happens. Every work order, every inspection, every maintenance activity that gets logged is training data for future decisions. FM teams that build this habit now are not doing administrative work. They are building the intelligence layer that AI will eventually use. Anyone wanting to start this journey can try out www.foxmy.io
3. Invest in people and technology. The talent shortage in Malaysian FM is real but it is also an opportunity. A structured FM team with clear competencies and career progression will extract more value from AI tools than an unstructured team with the same software. Technology does not replace skilled people. It makes skilled people significantly more effective.
4. Learn to measure value. FM teams that can quantify what they prevent, downtime avoided, asset life extended, energy saved, earn different conversations with their organisations. That discipline also makes AI outputs legible. If you cannot measure FM value manually, you will not trust what AI tells you either.
5. Build maintenance discipline and culture before buying predictive systems. Preventive culture before predictive. A structured maintenance programme with schedules and compliance tracking is not a stepping stone to be discarded when AI arrives. It is the foundation that makes predictive maintenance trustworthy.
Where AI Genuinely Earns Its Place in Facilities Management
Predictive maintenance works when there is clean historical data behind it. Condition monitoring works when there is a baseline to compare against. Automated prioritisation works when the underlying work order process is already structured. Energy optimisation works when consumption data is being captured consistently.
AI does not replace the discipline. It accelerates it.
The FM Teams That Will Lead This Transition
The FM teams in Malaysia that will lead this transition are not the ones waiting for the perfect moment to start. Implementing AI in facilities management in Malaysia needs a parallel strategy. They are the ones running both tracks right now. Imperfect data getting better. Processes getting more structured. Tools being added incrementally as the foundation strengthens.
The facilities management industry in Malaysia does not need to choose between fixing itself and embracing AI.
It needs to do both. At the same time. With the understanding that each one makes the other more valuable.
The conversation should not stop. It should evolve.
Start with the foundations. Build a complete asset register, record all operational activities systematically, and establish a structured preventive maintenance programme. These steps create the clean data that AI systems need to function effectively. Teams looking to begin can explore platforms like FOX at www.foxmy.io.
Parallel progress means fixing FM fundamentals and adopting AI at the same time rather than sequentially. It recognises that the industry cannot afford to pause on AI while it sorts out its basics, nor can it expect AI to work without structured data and processes underneath it.
Predictive maintenance requires clean historical data, consistent condition monitoring, and structured work order processes as its foundation. Most Malaysian FM teams are still operating reactively, which means the baseline data predictive systems need simply does not exist yet.
No. AI accelerates the work of skilled FM professionals rather than replacing them. A structured team with clear competencies and career progression will extract significantly more value from AI tools than an unstructured team with the same software. Investing in people and technology together is more effective than investing in either alone.
The FM teams that lead Malaysia’s transition will be those running two tracks simultaneously: building structured data and processes while incrementally adopting AI tools. The industry does not need to choose between fixing itself and embracing innovation. Both tracks make each other more valuable over time.