A strategy engagement, not a build: I ran discovery and stakeholder interviews across an airline's network-planning organization, scored its capabilities against where it needed to be, and delivered a sequenced recommendation set, a technology roadmap, and a TCO analysis with the economics of ROI for each option.
Confidentiality note. The client is anonymized and proprietary figures are withheld. Diagrams, scores and numbers shown here are sanitized recreations built for this portfolio, directionally accurate but not the client's actual data.
Network planning is one of the most complex processes in the business. Before anyone builds, the real question is where modernization earns its keep, in what order, and at what cost to own. That's what this engagement answered.
I ran discovery across the network-planning organization, tracing the process end to end — demand forecasting, route building, constrained demand, fleet assignment, and spill & recapture — through its commercial solver core. The goal wasn't to pick a tool. It was to separate the problems AI should touch from the ones it shouldn't, and to put a number on each.
Every candidate was run through a value assessment and the economics of ROI: the cost of the status quo, the realistic upside of modernizing, the delivery risk, and the payback period. That turned a wish list into a fundable, sequenced argument leadership could act on.
I interviewed more than twenty people across the function — leadership, revenue management, crew and maintenance operations, IT and data, and the frontline planners who run the schedule every day. The same themes surfaced from very different seats, which is how you know they're real.
"When demand moves, the re-plan takes days. By the time it's done, the world has moved again."
"Spill & recapture is the number the whole revenue plan rests on, and it's the slowest thing we run."
"I override the system by hand all day. It never learns from what I just did."
"The network lives in a dozen systems. Nobody has it as one connected picture."
"Whatever you change upstream, our constraints can't bend. That's non-negotiable."
"I'll fund it if someone can show me the payback. Nobody has put a number on it."
Five themes recurred: re-planning speed, trust in spill & recapture, tooling that ignores planner judgment, fragmented data, and unbending operational constraints — all underwritten by a finance team that wanted the ROI shown, not asserted.
I scored each capability on a five-point maturity scale — current state in solid green, target state as an open ring — so the gaps that mattered were obvious and the conversation moved from opinion to evidence.
The widest gaps — disruption re-planning, spill & recapture, and the data graph — lined up exactly with what stakeholders described. Constraint integrity was already strong and had to stay strong; nothing in the plan was allowed to weaken it.
Each recommendation carried an impact rating and an effort estimate, so leadership could fund the sequence rather than argue the parts.
One connected picture of every flight and airport that models can reason over — the foundation everything else depends on.
Reuse the base solve across thousands of scenarios in parallel, so a full-network re-evaluation runs in hours, not days — without bending a single constraint.
A forecast that adapts to planner overrides — a closed hub, a manual cut — instead of fighting them.
Surface the model's reasoning where planners already work, so judgment and automation reinforce each other.
Wrap and replace around the commercial solver rather than a risky big-bang rip-and-replace.
The roadmap sequenced the recommendations into Now, Next and Later — each horizon delivering value by itself, so momentum and budget never depended on the whole program landing at once.
Finance wanted the payback shown, not asserted. I modeled the five-year total cost of ownership for three paths — keep buying the commercial suite, rip-and-replace, or incrementally modernize around the core — including license, infrastructure, build, and the people to run it.
The recommended path wasn't the cheapest line item or the flashiest rebuild — it was the one with the best economics of ROI and the lowest delivery risk. Leadership funded it, and the spill & recapture engine at its center is now flying the network in production.