Services › Decision-Grade Modeling

6–8 week engagement · $30,000 – $60,000

Decision-Grade Modeling Brief

A six-to-eight week mechanistic, stochastic, or agent-based modeling engagement that answers one bounded question, co-developed with a named client-affiliated statistician.

When this is the right engagement

You have a decision in front of you and a generic model will not answer it.

What is the public health value of expanding airport surveillance to land borders? What happens to TB transmission if we shift from passive to active case-finding? Where should the next 50 community health workers be placed? Decisions like these deserve models built for the specific context — not retrofitted from a published paper.

What you get

  • A bounded, decision-grade model — mechanistic, agent-based, or compartmental, chosen to fit the question.
  • Open, reusable code in R or Python with a public repository.
  • A workshop deliverable delivered in-country/on-site where logistics allow, remotely otherwise.
  • Co-authorship on resulting peer-reviewed publications, with a named client-affiliated statistician as first or co-first author by default.

How the engagement unfolds

Weeks 1–2

Question framing & data audit

Define the decision, the estimand, and the data sources. Audit data quality and identify gaps. Issue a scoping memo with the modeling plan.

Weeks 3–5

Model build & validation

Build the model. Calibrate to local data. Run sensitivity analyses for the assumptions most likely to drive the decision.

Weeks 6–8

Workshop & deliverables

Deliver the workshop with the program team. Open-source the code. Submit a co-authored manuscript draft.

What this engagement will not do

This is not a black-box deliverable. Capacity transfer is part of the work, not an extra. If your team will not be involved in the build, this is not the right engagement. See Operating Principles.

Why this practitioner

Scenario modeling, built with the people closest to the decision.

  • NIH R01 PI · $2M, network-informed HIV prevention trial design
  • Mechanistic, agent-based, and compartmental models built in industry at Ginkgo Biosecurity and The Public Health Company
  • Co-author with Rwanda Biomedical Centre on the 2025 Kigali International Airport surveillance study (IJID Regions)
  • WHO Immunization consulting (2005, 2011); Harvard / De Gruttola HIV genetic linkage work in Botswana
  • Statistics Without Borders engagements with UNICEF and others