Specialists in Operational Efficiency

We redesign distribution networks, and prove the saving.

Smart Grid Systems models national supply chains at the level of individual roads, facilities and vehicles — then shows exactly which changes pay, by how much, and when.

See the evidence

Measured outcomes

The cost is rarely where people look for it

In one national medicines network, depot storage accounted for 78% of total spend. Transport accounted for 20%. Overnight stops, 2%.

Most optimisation effort goes into routing, because routing is visible and easy to measure. But in a network holding buffer stock across dozens of depots, the decisions that actually move the cost base are where stock sits and how much of it there is. Finding that out required modelling storage, transport and overnight costs separately rather than as one blended figure — and it changed what the client did next.

How we work →

Selected work

A national public-health medicines agency · Southern Africa

Redesigning a national medicines network

$14M saved annually, a 40% reduction

A last-mile redesign of a national public-health supply chain, establishing that depot siting and buffer-day policy — not routing — dominate network cost.

A national health supply-chain agency · East Africa

Rebuilding a national network model

8,877 health facilities modelled, nationwide

A national network-design study refreshing a decade-old baseline, covering 2.2 times as many facilities, with a desktop application handed over to re-run it.

A pharmaceutical distributor · South Africa

Last-mile delivery brought in house

R5.97M saved annually, around 35%

A two-site last-mile redesign that replaced an outsourced courier arrangement with a modelled in-house fleet, and set a defensible pace for electrification.

The same five stages, every time

What distinguishes a practice from a contractor is a method that repeats. Ours is deliberately conservative: it favours results that hold up under challenge over results that look impressive.

  1. Build a real road network. Real roads, with turn restrictions, tolls and measured speeds — not straight lines.
  2. Model demand and cost by activity. Storage, transport and overnight costs separated, because they answer to different decisions.
  3. Solve, rather than estimate. A constraint solver against stated constraints, so every recommendation has a reason behind it.
  4. Compare scenarios, not a single answer. A structured grid of scenarios, so the client sees the shape of the trade-off.
  5. Hand over the tooling. Where the engagement warrants it, the model itself — as an application the client can run.

The method in full →

Discuss a network