• Written by Muhammed Abdulla NC, CEO

In a meeting with the logistics team, the operations manager pointed out that despite implementing a new routing system, the actual cost savings were not materialising as expected. The team had assumed that by optimising routes, they would automatically reduce fuel consumption and lower costs. However, the reality was that carrier performance variance, dwell and detention, and exception handling at scale were all contributing to increased costs, eating into the expected savings.

Logistics complexities

The logistics team soon realised that carrier performance variance was a major contributor to the increased costs. Different carriers had varying levels of performance, and some were consistently late or failed to deliver on time. This variance made it difficult to predict and plan for costs, as the team had to factor in the potential for delays and additional expenses. Dwell and detention were also significant issues, as trucks were often forced to wait for extended periods at warehouses or distribution centres, incurring additional costs.

Exception handling at scale was another challenge the team faced. As the volume of shipments increased, so did the number of exceptions, such as damaged goods or missed deliveries. Handling these exceptions required significant resources and manpower, which added to the overall cost. The team had to develop strategies to mitigate these exceptions and reduce their impact on the bottom line.

Unit cost vs total cost

One of the key mistakes the logistics team made was focusing solely on unit cost, rather than total cost. They had assumed that by reducing the cost per unit, they would automatically reduce the total cost. However, this approach neglected to consider the impact of other factors, such as increased volume or changes in the supply chain. By using AsscherAi to analyse their data, the team was able to gain a deeper understanding of their total cost and identify areas where they could make adjustments to reduce costs.

The difference between unit cost and total cost is critical in logistics, where small changes can have a significant impact on the bottom line. By focusing on total cost, the team was able to identify opportunities to reduce costs that they may have otherwise missed. This approach also allowed them to make more informed decisions about investments in new technologies or processes.

Real levers for cost reduction

After re-evaluating their approach, the logistics team identified three key levers for reducing costs: caching, batching, and model routing. Caching involved storing frequently accessed data in a easily accessible location, reducing the need for repeated queries and minimising delays. Batching involved grouping similar tasks or shipments together, allowing the team to process them more efficiently and reduce costs. Model routing involved using advanced algorithms to optimise routes and reduce fuel consumption.

By implementing these strategies, the team was able to achieve significant cost savings. They were able to reduce the time spent on exception handling, minimise dwell and detention, and improve carrier performance. The team also developed a more nuanced understanding of their supply chain and was able to make data-driven decisions to drive further cost reductions.

Token cost

Token cost is often cited as a key factor in logistics costs, but in reality, it is rarely the dominant line. The logistics team found that while token cost was a significant expense, it was not the primary driver of costs. Instead, factors such as carrier performance variance, dwell and detention, and exception handling had a much greater impact on the bottom line. By focusing on these areas, the team was able to achieve more significant cost savings than they would have by simply reducing token cost.

This is not to say that token cost is unimportant, but rather that it should be considered in the context of the overall cost structure. By using AsscherAi to analyse their data, the team was able to gain a deeper understanding of their costs and identify areas where they could make adjustments to reduce expenses.

Limitations of AI cost optimisation

While AI cost optimisation can be a powerful tool for reducing costs in logistics, it is not a silver bullet. There are limitations to its effectiveness, and it is not always the right choice. For example, in situations where the supply chain is highly unpredictable or subject to significant disruptions, AI cost optimisation may not be able to provide accurate predictions or recommendations. In these cases, more traditional approaches to cost reduction may be more effective.

Additionally, AI cost optimisation requires high-quality data and significant computational resources. If the data is incomplete, inaccurate, or biased, the results of the optimisation process may be flawed. Similarly, if the computational resources are insufficient, the process may be slow or unable to handle complex scenarios. By understanding these limitations, logistics teams can make more informed decisions about when and how to use AI cost optimisation to drive cost reductions.