EN
Back to the archive

The encyclopedia · Strategy & Leadership · Operational decision · 2011–2013

Akshaya Patra's routing heuristic freed $1.96M a year for more meals

A three-stage clustering-routing heuristic cut Akshaya Patra's midday-meal distribution costs, projecting $1.96M annual savings and 2,400 more children fed.

Akshaya Patra Foundation

The solution

The Akshaya Patra Foundation runs midday meal programs for about 1.3 million children in more than 9,000 schools across nine Indian states, and food must arrive within a fixed time window before lunch.

Its distribution network had grown without a systematic plan. Researchers developed AMRUTA, a three-stage decomposition heuristic that clusters schools, assigns distribution vehicles to clusters and routes vehicles within each cluster.

A pilot in one location cut annual costs by US$75,000 — enough to add 2,400 children — and full implementation was projected to save about US$1.96 million a year.

Why it worked

  • Decomposition made a huge routing problem tractable with local tools.
  • The pilot proved savings before scale-up, de-risking adoption.
  • Every dollar saved feeds more children, so logistics is mission, not overhead.
  • The software approach was implementable in a developing-country nonprofit.
What it achievedCluster schools, then route; don't route all at onceinspired

What can be applied

When a network is too big to route at once, decompose it: cluster, assign, then route — each stage simple enough to implement, together a system a nonprofit can actually run.

Aftermath

AMRUTA's pilot results were published in Interfaces (2013) as a model for applying operations research in the social sector of a developing country.

Sources

spotted an error? The archive wants to know.

Related cases