Predictive Maintenance for a National Utility
Overview
Our client needed to stop discovering transformer faults from customer outage calls, and start seeing them coming weeks in advance.
Ian Mccarty
Chief Technology Officer / fairsystems
Utilities run assets that are expensive to replace, dangerous to ignore and largely invisible between inspections. Most still maintain them on a calendar rather than on evidence, which means servicing healthy equipment while failing equipment waits its turn. fairsystems was engaged by a national transmission and distribution operator to move its maintenance regime from fixed intervals to measured condition.
The Client
Our client operates high-voltage transmission and regional distribution networks serving several million connections. Their engineering teams are highly capable, but their asset data was scattered across a legacy asset register, handwritten inspection sheets and a SCADA historian that nobody outside the control room could query. They wanted to extend asset life, cut unplanned outages and build an evidence base they could defend to their regulator.
Roadmap
1
Assess
fairsystems began with a full asset criticality review, ranking every substation, transformer and switchgear unit by consequence of failure rather than by age. We interviewed control-room staff, field engineers and planners to understand where failures actually originate and how they are currently detected. In parallel our team audited the available data: twelve years of SCADA telemetry, the asset register, dissolved gas analysis records and the outage log. That audit established which failure modes were genuinely predictable from existing signals and which would need new instrumentation, so investment could be pointed at the gaps that mattered.
2
Deliver
Our consultants built a condition-scoring model for the asset classes that carried the highest risk, combining telemetry trends, oil analysis and load history into a single health index per unit. We integrated the model with the client's existing asset management system so planners saw health scores inside the tool they already used rather than in a separate dashboard. fairsystems specified and oversaw the installation of additional sensing on the small number of critical assets that lacked it, then reworked the maintenance planning process so work orders were raised from health scores and criticality rather than from the calendar.
3
Continue
fairsystems remains engaged on a quarterly cycle, retraining the condition models as new failure and intervention data accumulates and widening coverage to further asset classes. We work with the client's regulatory team to translate the health index into the evidence their price control submissions require, and with their planners to keep the intervention thresholds tuned as the asset base ages. Our consultants also run an annual review of instrumentation coverage so that new sensing is added where the models are least confident, rather than uniformly across the network.
Solution Details
Assets fail on their own schedule. Maintenance should follow evidence, not the calendar.
Asset condition data scattered across incompatible systems
Deliverable: fairsystems consolidated the asset register, SCADA historian and inspection records into a single asset data model, giving every unit one authoritative condition history.
No objective way to rank which assets to intervene on first
Deliverable: Our consultants built a criticality-weighted health index that scores every asset on both probability and consequence of failure, so intervention budgets go to the units that matter.
Maintenance planned on fixed intervals regardless of condition
Deliverable: We redesigned the planning process so work orders are generated from condition thresholds, retiring the blanket calendar cycle for the asset classes where evidence supports it.
Critical assets with insufficient instrumentation
Deliverable: fairsystems ran a gap analysis against the model's confidence scores and specified targeted sensing only where it measurably improved prediction, avoiding blanket sensor spend.
Field engineers sceptical of model-driven work orders
Deliverable: Our team built the health index to be explainable, showing the contributing signals behind every score, and ran hands-on sessions so engineers could challenge and correct it.
Regulatory submissions unsupported by asset evidence
Deliverable: We mapped the condition model outputs to the evidence categories the regulator expects, turning the health index into a reusable input for price control submissions.
Spare parts held against the wrong assets
Deliverable: fairsystems re-based the critical spares policy on predicted intervention demand rather than historical consumption, releasing capital held in slow-moving stock.
No feedback loop from intervention outcomes back to the model
Deliverable: Our consultants instrumented the work order process so every intervention records what was actually found, feeding the result back into the next model retraining cycle.
Result:
The client now plans maintenance from measured condition rather than elapsed time. Unplanned outages on the covered asset classes fell substantially in the first full year, and interventions are concentrated on the units that genuinely need them. Engineers trust the health index because they can see what drives it, and the regulatory team has an evidence base it can defend rather than reconstruct. Most importantly, the client owns the model: their own analysts retrain it, extend it and challenge it without depending on us to do so.
0
Unplanned outage reduction
0
Maintenance cost reduction
0
Asset life extension
0
Critical spares capital released
0
Assets under condition monitoring
Testimonials
How our services bring about success
Annika Nkosi
Lead Process Analyst
fairsystems’ team brought a unique combination of skills to our business assignment and was often sought out to tackle the most complex and integrated challenges.
Marie-Anne Badeaux
Programme Manager
fairsystems will challenge your thinking, provide you with alternative perspectives and, where appropriate, actively coach/mentor/advise you to generate and co-create solutions.
Justin J. Spratt
General Manager
Fairsystems consults impeccably to understand organizational needs and then brings innovative business relevant ideas and solutions to the table.
Mohamed Bin Naser
Chief Financial Officer
I am impressed with fairsystems’ team ability to grasp complex, accounting system problems and quickly find a simple, straightforward and affordable approach.
Ella-May Robinson
HR Manager
fairsystems has an incredible gift for making the complex appear simple and without being simplistic. It is a pleasure to work with them as they always deliver to an exceptionally high standard.
Jamal Aagha
Chief Executive Officer
There are some great companies in the media industry and the social space, but we have yet to see anyone take it to the level that we have with @fairsystems.
Allen Gibson
Operations Manager
The most valuable lesson we learned is that there are no shortcuts. Every minute we spent dissecting and learning was rewarded.
