Food Technology / Supply Chain
·Singapore & Southeast Asia·7 monthsAI Supply Chain Matching and Logistics for a Food Rescue Platform
Deployed AI-powered supply chain matching and logistics optimisation for a food rescue platform that connects surplus food from suppliers with buyers at reduced prices. The system automates supplier-buyer matching based on product type, quantity, location, and time-sensitivity, optimises delivery routing for perishable goods, and predicts surplus availability to enable proactive buyer engagement.
Minutes
Matching time (was hours)
35%
Reduction in delivery cost per transaction
3x
Transaction volume increase without headcount growth
Challenge
Food rescue platforms operate at the intersection of supply chain logistics and social impact, dealing with a uniquely time-sensitive matching problem. This platform connected food suppliers — manufacturers, distributors, restaurants, and retailers — with buyers willing to purchase surplus food at discounted prices, diverting it from landfill. The core challenge was matching: surplus food has a limited shelf life, varies in type and quantity unpredictably, and must reach a buyer who wants that specific product in that specific quantity within a narrow time window. The matching process was largely manual. When a supplier listed surplus inventory, the platform's operations team would identify potential buyers from their network, contact them to gauge interest, negotiate pricing, and arrange logistics. This manual process worked at small scale but became a bottleneck as the platform grew. Response time was critical — a pallet of dairy products with three days of shelf life needs a buyer within hours, not days. Every hour of delay increased the probability that the food would expire before reaching the buyer, converting a revenue-generating transaction into waste. Logistics coordination was the other major pain point. The platform used a combination of its own delivery fleet and third-party logistics providers. Routing perishable goods efficiently — accounting for cold chain requirements, delivery time windows, traffic patterns, and vehicle capacity — was beyond the operations team's ability to optimise manually. Suboptimal routing led to higher delivery costs (which reduced the price advantage that motivated buyers) and longer transit times (which consumed precious shelf life).
Approach
Reluvate built an intelligent matching and logistics platform. The matching engine operates in real-time: when a supplier lists surplus inventory, the system immediately identifies potential buyers based on historical purchase patterns, product preferences, location proximity, purchasing frequency, and available delivery capacity. Matched buyers receive instant notifications with product details and a time-limited offer. The system handles multi-party matching for large surplus lots — splitting a pallet across multiple buyers based on each buyer's capacity and preferences. The logistics optimisation module generates delivery routes that minimise transit time (preserving shelf life) while maximising vehicle utilisation (reducing per-unit delivery cost). The routing algorithm accounts for cold chain requirements (which vehicles have refrigeration), delivery time windows specified by buyers, real-time traffic data, and the perishable nature of the cargo — prioritising deliveries for items with the shortest remaining shelf life. Routes are dynamically adjusted as new orders are matched and existing deliveries are completed. The predictive module analyses historical surplus patterns — by supplier, product category, day of week, and season — to forecast when and where surplus will become available. This enables proactive buyer engagement: the system can notify high-probability buyers before surplus is listed, priming demand and reducing the time between listing and matching. For suppliers, the system provides waste reduction analytics showing how much inventory was diverted from landfill and the associated cost savings and sustainability impact.
Design Notes
The matching algorithm was designed to balance multiple competing objectives: speed (matching must happen within minutes, not hours), price optimisation (the platform's commission depends on transaction value), logistics efficiency (nearby matches are cheaper to deliver), and platform health (concentrating all matches on a few large buyers would create dependency risk). Reluvate worked with the platform's team to define an explicit objective function that weighted these factors, with parameters that could be adjusted as business priorities evolved. Change management was minimal because the system augmented an obviously bottlenecked manual process. The operations team had been working at capacity, turning away potential suppliers because they couldn't process matches fast enough. The AI system immediately increased their capacity by an order of magnitude, and the team's role shifted from manual matching to exception handling (unusual products, first-time buyers, quality concerns) and strategic relationship management. Exception handling addresses the perishability constraint directly. If a match fails — buyer cancels, delivery delayed, quality issue on inspection — the system immediately re-matches the surplus with alternative buyers, recalculating logistics in real-time. For items approaching their shelf life limit without a match, the system progressively widens the matching criteria (larger geographic radius, lower price thresholds) and eventually triggers donation routing to food banks as a zero-waste fallback.
Result
Average time from surplus listing to buyer matching dropped from hours to minutes. Delivery cost per transaction decreased through optimised routing. The platform's transaction volume scaled significantly without proportional growth in operations headcount. Food waste diversion metrics improved measurably as faster matching and proactive demand priming ensured more surplus found buyers before shelf life expiration. The predictive surplus model enabled the platform to offer suppliers proactive surplus management, strengthening supplier retention.
Related Case Studies
Retail
Autonomous Operations for a National Furniture Retailer
Deployed end-to-end autonomous operations across five business functions — HR, procurement, customer service, accounts, and marketing — for a national furniture retailer with multiple showrooms. This is Reluvate's longest and deepest client engagement, running continuously for over 2.5 years with expanding scope.
Events & Hospitality
AI Customer Engagement Platform for a National Events Company
Deployed a multi-channel AI chatbot for a national events company, handling attendee inquiries, event registration, vendor coordination, and post-event follow-up across WhatsApp, web chat, and social media channels. The system reduced customer service headcount requirements during peak event periods while improving response times from hours to seconds.
Retail & Conservation
Computer Vision Retail Analytics and Wildlife Monitoring
Developed computer vision systems for two distinct domains: retail analytics (footfall tracking, customer heatmaps, and employee activity monitoring across store networks) and animal welfare monitoring (behavioural analysis, temperature tracking, and health anomaly detection). The retail system outperformed a major Chinese AI competitor's solution in accuracy benchmarks.