• Written by Muhammed Abdulla NC, CEO

The objection that cost optimisation in automotive is too complex to be tackled by AI is misguided, as it misunderstands the nature of the problem and the capabilities of AI systems like AsscherAi, which can provide real-time answers to queries about an organisation's own data, including production and supply chain information, and help identify areas for cost reduction, as can be seen on our website.

Automotive industry challenges

The automotive industry presents unique challenges for cost optimisation, including supplier escalations, engineering change propagation, and warranty root cause analysis, which can have a significant impact on production costs and timelines. For instance, a change in the design of a component can trigger a cascade of changes throughout the supply chain, leading to delays and increased costs. A production manager at a major automotive manufacturer, for example, may need to track the impact of these changes on the production line and adjust schedules accordingly.

Furthermore, the tiered supply chain in the automotive industry, with its tight takt and long programmes, adds an extra layer of complexity to cost optimisation efforts, as changes in one part of the supply chain can have far-reaching consequences. This complexity makes it difficult to identify areas for cost reduction and to implement changes that will have a lasting impact.

Unit cost vs total cost

A common mistake in cost optimisation efforts is to focus solely on unit cost, without considering the total cost of a given workflow or process. In the context of AI cost optimisation, this means looking beyond the cost of individual components or processes and considering the overall cost of the workflow, including factors such as labour, materials, and equipment. A shift supervisor at a manufacturing plant, for example, may be able to reduce the unit cost of a component by sourcing it from a cheaper supplier, but if this change leads to increased labour costs or equipment downtime, the total cost of the workflow may actually increase.

This highlights the need for a more holistic approach to cost optimisation, one that takes into account the complex inter relationships between different parts of the workflow and the supply chain. By using AI to analyse data from across the organisation, companies can gain a deeper understanding of where costs are incurred and identify opportunities for reduction that may not be immediately apparent.

Levers for cost reduction

There are three key levers that can be used to reduce costs in AI workflows: caching, batching, and model routing. Caching involves storing frequently-used data or results in a easily-accessible location, reducing the need for repeated calculations or queries. Batching involves grouping similar tasks or processes together, allowing for more efficient use of resources and reducing waste. Model routing, on the other hand, involves optimising the flow of data and tasks through the workflow, reducing unnecessary steps and improving overall efficiency. By using these levers, companies can reduce costs and improve the overall performance of their AI workflows.

For example, a category manager at a automotive manufacturer may be able to use caching to store information about supplier performance, allowing for faster and more accurate analysis of supplier data. By batching similar tasks together, the category manager may be able to reduce the time and resources required to complete these tasks, freeing up staff to focus on higher-value activities.

Token cost

Token cost is often cited as a major factor in the cost of AI workflows, but in reality it is rarely the dominant line item. While token cost can be significant, it is often dwarfed by other costs such as labour, equipment, and materials. Furthermore, token cost is often a fixed cost, incurred regardless of the actual usage of the AI system, whereas other costs may be variable, depending on the specific workflow or process being implemented. A more nuanced understanding of the costs involved in AI workflows is needed, one that takes into account the complex inter relationships between different cost factors.

This is not to say that token cost is unimportant, but rather that it should be considered in the context of the overall cost of the workflow, rather than being viewed as the sole determinant of cost. By using AI to analyse data from across the organisation, companies can gain a deeper understanding of where costs are incurred and identify opportunities for reduction that may not be immediately apparent, and contact us to learn more about how AsscherAi can help.

Limitations of AI cost optimisation

While AI cost optimisation can be a powerful tool for reducing costs and improving efficiency, it is not a panacea for all cost-related problems. There are certain situations in which AI cost optimisation may not be the best choice, such as when the workflow is highly variable or unpredictable, or when the costs involved are largely fixed and unchanging. In these cases, other approaches to cost optimisation may be more effective, such as traditional cost accounting or process improvement methodologies.

Additionally, AI cost optimisation is not a replacement for human judgement and expertise, but rather a tool to support and augment these capabilities. While AI can provide valuable insights and recommendations, it is still up to human decision-makers to interpret and act on this information, taking into account the broader business context and strategic goals of the organisation.