Secure the Harvest — Rima Foods and the Empty Gate

A four-round, advanced Operations & Supply-Chain simulation set inside Rima Foods Limited, a tomato triple-concentrate plant in Sokoto State, Nigeria, rated for 900 tonnes of fresh fruit a day and averaging 387 (43% utilisation). Around 4,200 smallholders farm within a 60 km radius on an average 1.8 hectares. The plant was built in the growing area because the roads south make fresh tomato unmovable without loss, and its buying policy has never changed: a fixed posted price per tonne for the whole season, paid at the factory gate within 48 hours, with nobody from Rima going out into the market. Tomatoes in Nigeria are traded by the raffia basket — 40 to 80 kg depending on how it is woven — at a price that moves every morning; last season the open market ran ₦118,000–₦310,000 per tonne against a posted ₦180,000. The consequence is the premise: the plant fills when the market pays less than Rima and empties the moment it pays more, and 61% of intake in the expensive weeks arrives after 17:00 — produce that did not sell and would spoil overnight. Rima is not a buyer; it is the farmer's option of last resort. Playing the Supply Chain Director across four growing seasons, you (1) READ THE MARKET — name the root cause as the buying mechanism rather than the price level, decide whether to wait at the gate or put buyers into the markets, choose whether to quote in kilograms, in baskets or in both with a published conversion, and set a posted-price premium knowing fresh fruit is two-thirds of the cost of a tonne of concentrate, so every point is multiplied by six before it reaches the P&L; (2) CHOOSE THE SUPPLY MODEL — independent smallholders, contract farming with seed, agrochemicals, irrigation and extension, a joint venture with a registered cooperative that takes nine to eighteen months to incorporate, a lease, or your own plantation acquired through the state, against a participation constraint that caps the crop you can sign at roughly 85×(1−e^(−premium/11)) per cent because nobody signs a contract for nothing; (3) SURVIVE THE CYCLE — a pest outbreak cuts regional output by a third and pushes the market to four times your posted price, and you choose whether to enforce the contract, renegotiate upward unprompted, release farmers to the market or penalise the defaulters, with the retention consequence landing a season later; and (4) MAKE THE PLANT PAY — run to a committed schedule, stay in batch, add a second line or add cold storage, position on traceability or head-on against imported Chinese paste, and defend a three-year supply plan. Cost per tonne is derived, not assumed: farm-gate price through the 6:1 conversion, plus farmer support, plus ₦4.85bn of season fixed cost recovered over however much concentrate the plant actually made — so utilisation moves cost, and a premium that lifts utilisation can still lose money. The scoring composite weights cumulative EBIT (30%) and farmer retention (24%) above utilisation (16%) on purpose: a team can drive the plant to 95% and finish poorer than one that ran it at 84%. The model punishes the five classic errors: buying volume with a premium instead of fixing the mechanism, contracting the crop while putting nothing into the farmer, enforcing contracts punitively during the shock season, taking the land route and ignoring the community and political risk it carries, and adding a second processing line while the first runs half empty. Final KPIs track Plant Utilisation (%), Farmer Retention (%), Cost per Tonne of concentrate (₦'000) and cumulative EBIT (₦bn).