Sensorry.
Design the network, not just the sensor.
Sensorry packages the multi-objective sensor-optimisation workflow from my doctoral research — MOSOF and the Normalised Diagnostic Contribution Index — into a tool that engineering teams can use directly. Where AcoustR ships the diagnostic, Sensorry ships the design step that comes before it: which sensors, where, and at what cost-coverage trade-off.
Sensor decisions are made wrong.
On most complex assets — aircraft systems, industrial plants, vehicles — sensor networks are specified by accumulation. Each subsystem owner adds the sensors they need, no one owns the whole picture, and the result is a stack that's simultaneously over-instrumented in some places and blind in others.
Sensorry is built on a different premise: sensor selection is an optimisation problem with measurable objectives — diagnostic coverage, cost, weight, reliability — and the right answer is a Pareto front, not a wishlist.
Candour about the stage
Sensorry is currently a working prototype, not a generally-available product. The methods it implements are peer-reviewed and validated; the productisation around them — UI, data ingest, customer onboarding — is the work in progress. If you'd be a useful early partner, the contact form below is the path.
Three stages, one front.
Encode the system
Define the asset, candidate sensor positions, failure modes of interest, and the objectives the team actually argues about — cost, weight, reliability, coverage.
Search the space
A multi-objective genetic algorithm searches the candidate-network space and scores every configuration with NDCI for diagnostic contribution.
Deliver the front
The team gets the Pareto front and a knee-point recommendation — with explicit trade-offs, not a single black-box answer.
The thesis, made usable.
The methods Sensorry implements were developed and validated across four aircraft subsystems (Engine, Fuel, EPS, ECS) through Cranfield's SESAC platform, with three first-author publications in Sensors (MDPI). The published results consistently identify compact Pareto-efficient sensor configurations that match or exceed the diagnostic capability of larger reference designs.
Sensorry exists to make those results accessible to teams who don't want to re-implement the framework from a paper. It's the same idea as Sensorry's sister venture AcoustR — research that doesn't get used isn't research that matters — applied one layer up the stack.
Designing a network right now?
If you're scoping a sensor network on a complex asset and you'd value an external view — or you'd consider being an early Sensorry partner — get in touch. The product isn't general-availability yet, but the methods are ready to apply to your problem now.
hello@buraksuslu.com →