Innovating from Jos: Oyebode speaks on data, discipline, Nigeria’s power future

By Meyaptong Gyang

An engineer, Oyegoke Oyebode, who is innovating from Jos, Plateau State has spoken on data, discipline and Nigeria’s power future.

In February 2021, electricity once again dominated national debate. Tariff hikes sparked public anger across Plateau State and the wider North Central region.

In Jos, the frustration was familiar: higher bills paired with unreliable supply. Yet beneath the noise, reforms first introduced in 2018 were quietly reshaping Jos Electricity Distribution Plc (JED Plc).

Among the engineers behind those reforms was Oyebode, a graduate of the University of Lagos (UNILAG), who spoke to ThisDay with a simple conviction: no system improves without data and structure.

You were in Jos a few years before 2021. What was the reality you encountered?

The real issue wasn’t just broken infrastructure—it was broken information. Procurement cycles dragged on, warehouses overflowed with the wrong parts, and engineers often lacked the essentials to fix faults in the field. Managers leaned more on instinct than on evidence, and customers paid the price through endless outages.
We set out to change that.

By embedding measurement into daily operations and linking procurement to actual consumption patterns, we built systems that could anticipate needs instead of scrambling after the fact. It was about shifting from improvisation to prediction—an early step toward what I now think of as intelligence-led design.

How did those changes translate into measurable impact?

The impact showed up where it mattered most—on the grid and for customers. Outages that once stretched for days could now be restored in hours, because the right materials were finally reaching engineers when they were needed. Fault recurrence dropped, so neighborhoods weren’t plunged back into darkness after a repair.

Feeder reliability also improved. Average restoration times fell by about 4%, a small but meaningful shift that signaled a move toward faster, more predictable service. For households and businesses, that meant less disruption and greater confidence in the grid. For the first time, reliability was being engineered into daily operations rather than left to chance.

How did those lessons carry into your research on Artificial Intelligence (AI)?

Jos taught me that inefficiency isn’t random—it’s patterned. Outages weren’t accidents; they were predictable. Procurement delays weren’t mere bureaucracy; they were optimization problems waiting to be solved. Outages aren’t accidents, they are predictable.

Later, when I moved into AI research, those same frustrations became the foundation of my models. Forecasting tools began predicting shortages before they happened. Machine learning flagged equipment stress before it triggered blackouts. Optimization techniques aligned resources with demand more precisely.

Everyday failures became training data. By reframing them as structured problems, we could design systems that prevented disruption rather than simply reacting to it.

And when you moved on to Xerox, what changed? The setting was different, but the questions were familiar: how do you make complex systems reliable, efficient, and cost-effective?

At Xerox, we applied predictive intelligence and automation in smart ways. Real-time anomaly detection flagged issues before breakdowns. Automated troubleshooting cut diagnostic times. Smarter logistics ensured that parts and engineers got where they were needed with less waste.

The outcomes were tangible with fewer delays, greater reliability, and lower costs. But the lasting lesson was continuity: whether in Jos or at Xerox, inefficiency is not an accident. It’s a system flaw. And once you recognize it as such, you can design it out.

What connects those two experiences?

For me, the thread is discipline. In Jos, discipline meant data visibility. At Xerox, it meant predictive control. Different contexts, same principle: systems don’t improve through improvisation, they improve through structure. Systems don’t improve through improvisation, they improve through structure.

When you treat inefficiency not as a nuisance but as a pattern to be modeled, you can design processes that last. That’s what keeps the lights on, literally in Jos, and figuratively in every other system I’ve worked on since.

Nigeria’s grid is infamous for being reactive rather than predictive. What would it take, technically and institutionally to embed real-time intelligence at scale?

The first step is visibility. You can’t predict what you don’t measure. That means deploying smart meters, grid sensors, and automated monitoring across transmission and distribution—not as pilot projects, but as the norm.
But technology alone won’t fix it. The institutional structure has to change.

Today, generation companies, transmission, and distribution all operate in silos. A predictive grid requires integration—one digital backbone where data is trusted, shared, and acted on in real time.

With that foundation, outages stop being “surprises.” They become forecastable events. Maintenance shifts from emergency repairs to preventive action. The system evolves from firefighting into optimization. The barrier isn’t technology, it’s discipline. And discipline begins with data.

Looking ahead, can Nigeria leapfrog straight into AI-driven distributed grids, bypassing the old centralized model, the way we skipped landlines with mobile phones?

Absolutely and it may be our most practical path. Reforming the centralized grid is capital-intensive and politically slow. Distributed systems—mini-grids, solar clusters, hybrid networks—are faster, cheaper, and more adaptable.
The parallel with telecoms is striking. We didn’t wait for copper landlines in every home; we jumped straight to mobile. The same can happen with power.

We didn’t wait for copper landlines; we jumped straight to mobile. Power can follow the same path. If predictive intelligence is embedded into distributed grids from day one through demand forecasting, anomaly detection, AI-optimized storage, then reliability can grow from the edges inward.

The national grid will still matter, but its role will shift. Instead of being the sole backbone, it will act as a stabilizer while innovation flourishes at the periphery. That’s how Nigeria can leapfrog—not by copying outdated models, but by designing an intelligent system suited to our reality today.

Conclusion: A Future Anchored in Systems

The challenges of Jos’s grid in 2018 weren’t solved overnight. But the approach pioneered there, data discipline, predictive thinking, and structured engineering, offer a glimpse of what reform can look like.

For Nigeria, the lesson is clear: just as roads and bridges carry goods, data systems and predictive frameworks carry reliability. Without them, no investment in power will endure.

Progress begins with engineers who measure what others ignore, translate failures into data, and design processes where improvisation once ruled. That is how reliability is engineered. That is how trust is restored. And that is how Nigeria’s energy future can be secured.