Skip to content
Dubai MunicipalityGovernmentUAE2026

Three modules that read a feasibility study, judge it, and write a new one from scratch

Not document automation. A feasibility platform that does the analyst's entire job — and then does the job the analyst could never do at all.

The challenge

Feasibility assessment inside a municipal authority means working through large volumes of dense technical and regulatory documentation, where the material that determines a decision is buried across hundreds of pages — and arrives in whatever format the submitting consultancy chose to use.

The work does not stop at reading. Every assessment produces downstream artefacts: financial sheets, standardised forms, calculation tables. These were being filled by hand, transcribing figures out of one document and into another, which is slow and exactly the kind of task where transcription errors quietly enter the record.

Benchmarking was harder still. Judging a proposal against ISO, WHO and other international standards requires knowing which clauses apply and finding the corresponding evidence in the submission — expert work that varied depending on who performed it.

And commissioning a feasibility from scratch meant going outside: a professional study is a six-to-twelve-week external consultancy engagement, even when the authority already holds the historical data the study would be built from.

Our approach
01

Module 01 — Analyse and populate

Ingests any feasibility study, external or internal, parsing inconsistent structures into a common representation rather than requiring a fixed template. Every figure is extracted and the complete Excel model is filled — sheets, formulas and calculation chains computed from the source document rather than retyped from it.

02

Module 02 — Benchmark and gap-analyse

Scores the submission against ISO, WHO and other international frameworks, and returns a full gap analysis: what is met, what is unevidenced, what falls short, and what it would take to close each gap. The same judgement applied the same way, regardless of who is reviewing.

03

Module 03 — Generate from scratch

Produces an entirely new professional feasibility study from the authority's own local historical data and current market conditions, written to Big Four consulting standards. Modules 01 and 02 make an existing process faster; this one removes the need to commission the study externally at all.

04

Structured for review, not blind acceptance

Every output is built for a reviewer to interrogate — the extracted values trace back to their source, the benchmark shows its evidence, and the generated study reads as a draft to be signed off rather than an answer to be trusted.

Source · third-party feasibility
Flagship · Dubai Municipality

It reads the study.
Then does the work after it.

Feasibilities read, forms and financial sheets filled and calculated from the source, then benchmarked against ISO and WHO standards.

Financial sheet · auto-populated
  • B4Capital cost48,200,000
  • B5Operating cost3,100,000
  • B6Discount rate6.0%
  • B7NPV =NPV(B6,B4:B5)38,940,220
Benchmarked against world standards
  • ISO 31000
  • WHO air quality
  • ISO 14001
What we built

The system, in specifics.

AI extraction from third-party feasibility studies and internal feasibility documents, handling varied formats and structures.

Automated population of the forms and sheets each feasibility requires, driven by the extracted data.

Financial sheet completion with formulas applied, so calculations derive from the source rather than manual transcription.

Feasibility analysis benchmarked against international standards including ISO, WHO and other recognised frameworks.

Full gap analysis per submission: requirements met, requirements unevidenced, and what closing each gap would take.

Generation of an entirely new professional feasibility study from local historical data and current market conditions, to Big Four consulting standards.

Automated summarisation of critical sections within large regulatory and technical filings.

Standardised feasibility study generation, removing structural variance between assessments.

Outcome

What changed.

Extraction from both external consultancy submissions and internal documents

Forms and financial sheets populated and calculated automatically from source

Transcription errors removed from the calculation chain

Feasibilities benchmarked consistently against ISO, WHO and other world standards, with a gap analysis per submission

A complete professional feasibility generated in-house from historical data, without commissioning an external consultancy

Standardised reporting structure across all submissions

Delivered for a government authority

Next

Related work.

AI Construction Estimator — Construction takeoff and bidding from architectural PDFs, in minutes
Flagship

AI Construction Estimator

Multi-client programmeConstructionUSA

Construction takeoff and bidding from architectural PDFs, in minutes

Reads architectural PDFs, detects the scale on its own, measures only the trade you work in, and returns quantities, labour hours and a bid — then pushes the work into your CRM.

Computer VisionDocument AIDeep Learning
Read case study
Flagship

Moving Estimator

Umzugsauktion GmbH & Co. KGLogisticsGermany

A walkthrough video in, a removals quote in seconds

The customer films their own home. Computer vision identifies and counts every object, estimates dismantling, handling and transport time, and returns a priced quote.

Computer VisionObject DetectionDeep Learning
Read case study
Flagship

Outcall Review Agent

Multi-client programmeEnterpriseGlobal

It calls every customer, then tells the board what they said

An AI agent calls every customer after service and gathers the review on the call itself. Every conversation is then analysed automatically and clustered into a picture of what is happening per department, per individual and across the whole organisation.

Voice AILLMsSpeech Recognition
Read case study