The challenge
Transport planners operate daily at the intersection of time, capacity and quality. Deliveries must be made on time, while a mix of controllable and uncontrollable factors comes together in a single plan. It is precisely this accumulation of variables that makes determining the optimal route complex.
In the dairy industry, this complexity is further increased by the nature of the product. Fresh milk and other perishable raw materials require tight time windows and precise coordination. Fat and protein levels vary by supplier, volumes differ per location, and specific contractual agreements determine where raw materials can be processed. The core question was how to transport fresh dairy from dozens of farms to the right processing facility, via the right routes, without loss of quality and at the lowest possible cost.
In practice, planning was largely done manually, using extensive Excel files as a foundation, supported by experience and intuition. This made the process labour-intensive and prone to error, while the perishability and bacterial sensitivity of the product increased the risk. If a tanker was on the road for too long, an entire load could be rejected, with immediate financial impact.
There was a clear need for a scalable, data-driven solution that combines speed, accuracy and stronger decision-making.
The approach
Together with the organisation, KPMG developed an AI-driven planning assistant that leverages existing data on transport orders, fleet composition, driver availability and product specifications. Based on this data, an optimal plan is generated within a short time frame.
At its core is a constraint-based optimisation model, in which multiple conditions jointly determine the outcome. Thousands of variables, including routes, volumes, contractual agreements, fat and protein levels and available capacity, are calculated simultaneously and translated into a single plan that meets all operational requirements.
The cloud-based solution uses an advanced optimisation engine and can be continuously updated with new information, such as traffic conditions, changes in product composition or a vehicle breakdown. This creates a dynamic system that not only optimises, but also adapts to the reality of daily operations.
For planners, this marks a clear shift in responsibilities. Less time is spent on manual puzzle-solving, with more focus on oversight, exception handling and strategic decision-making.
The initial optimisation focused on efficiency and immediately delivered savings of several million euros. At the same time, it created room to explicitly incorporate additional objectives, such as reducing CO₂ emissions or optimising driver utilisation. The organisation’s well-structured data foundation proved to be a key enabler in this process.
The result
With this approach, the organisation now has a scalable solution that structurally improves the collection and delivery of raw materials. Continuous optimisation based on real-time data leads to lower costs, reduced waste and better-informed decisions, while also laying a strong foundation for further digitalisation and the integration of sustainability goals within the transport chain.
The team that made the difference
A team of AI and optimisation specialists from KPMG developed the planning assistant and underlying optimisation model together with the organisation, combining technological expertise with deep logistics insight to deliver a solution that works in daily practice.
