Order Ops
Order Ops is a six-player ecosystem simulation of online food delivery platform operations. Two participants play as competing Platforms and four as Restaurants, each operating individually within a shared system. Platforms negotiate commission rates and penalty structures with restaurants, manage order acceptance, assign delivery drivers from a shared pool, and estimate delivery times. Restaurants accept or reject orders based on kitchen capacity, estimate preparation times, and prioritise order queues. Performance scores — the Platform Performance Score and Restaurant Performance Score — drive order allocation through a Softmax function, creating a direct feedback loop between operational precision and market positioning.
The simulation runs across four phases: a negotiation phase, normal Day 1 operations, a Day 2 rain disruption scenario that reduces driver availability by 30% while sharply increasing order volumes, and a final recovery phase. The shared driver pool lies at the heart of the simulation's design — it creates interdependencies across all six players that make individual optimisation insufficient, and reveals how small operational failures cascade rapidly across a three-sided market.
- Coordinate customers, restaurants, and drivers
- Negotiate commissions and penalty structures
- Manage capacity under demand uncertainty
- Analyse performance-driven allocation dynamics
- Optimise delivery estimates under uncertainty
- Fixed six-player ecosystem (2 Platforms, 4 Restaurants) with distinct roles, decision sets, and performance metrics for each side
- Negotiation phase where platforms propose commission rates (10–35%) and tiered penalty structures, with no-agreement defaults that disadvantage both sides
- Real-time statistics dashboards displaying market share, Platform and Restaurant Performance Scores, profits, penalties, driver utilisation, and on-time delivery rates
- Day 2 rain disruption scenario with a 30% reduction in driver availability and a sharp increase in order volumes, exposing shared resource constraints
- Fully customisable parameters including order generation rates, driver pool size, kitchen capacities, Softmax sensitivity, and disruption settings
- Understand how three-sided platform markets create interdependencies that make individual optimisation insufficient
- Recognise how performance score feedback loops amplify small operational failures into significant market share losses
- Appreciate that negotiated commission and penalty structures function as binding constraints, not merely opening positions
- Experience firsthand how shared resource constraints — particularly a finite driver pool — create tragedy-of-the-commons dynamics under demand surges
- Develop the ability to balance aggressive commercial targets with ecosystem-level cooperation across platforms, restaurants, and delivery capacity
