Agrisight

Context and objectives

Cereal collectors and traders face costly uncertainty on their margins, as both harvests and sales are hard to predict.

Given the complexity of the supply chain (many products, storage points, clients and transport modes), existing planning capabilities and tools are limited, making it difficult to update plans when assumptions suddenly change.

Outcome

1. A 15 to 20% reduction in logistics costs
2. A 15% reduction in CO2 emissions
3. Much easier planning processes

Our approach

Step 1: a digital twin

• A single place to manage all supply chain data, including data held in no ERP or other system (e.g. silo characteristics)
• Integration of commercial data, to measure the impact of any decision on sales margins

Step 2: a long-term planning module

• An advanced optimisation algorithm that proposes the best plan, taking into account hundreds of constraints and parameters, with a focus on cost and CO2 reduction
• An interface that avoids the black-box effect and lets users force certain constraints or logistics flows to reflect special situations
• A new plan or scenario in minutes

Step 3: harvest and sales predictions feeding the planning tool

• A user-friendly interface to update human forecasts quickly, combined with AI that factors in historical data, client data (e.g. progress on contract execution) and external data
• Shared use by supply chain and commercial teams