• Written by Muhammed Abdulla NC, CEO

A meeting where the supply chain team presents a new tool that can answer any question about inventory and delivery, but the team leader barely looks up from their notes, is not uncommon. The tool is demonstrated, with examples of how it can provide real-time answers to complex queries, but the interest is lukewarm at best. This is often the fate of new enterprise AI systems, including those designed for supply chain management, which are capable of analysing vast amounts of data but end up being underutilized. The reason for this is not the technology itself, but how it is integrated into existing workflows.

Attaching new capability to a routine that already exists

The key to successful adoption of enterprise AI for supply chain management is to attach the new capability to a routine that already exists. This means identifying a specific task or process that the team is already performing, and showing how the AI system can enhance or automate that task. For example, a shift supervisor checks the maintenance log every morning to plan the day's work, and the AI system can provide real-time updates on inventory levels and delivery schedules to inform that planning. By building on an existing routine, the team is more likely to adopt the new technology.

This approach also helps to build trust in the AI system, as the team can see the direct impact it has on their daily work. It is not about replacing existing processes, but about augmenting them with new capabilities. A category manager compares two suppliers, not just on price, but on lead time and inventory availability, using the AI system to provide the necessary data.

Who needs to be able to ask, not just who needs the answer

Most people get wrong the idea that only a select few need to be able to ask questions of the AI system. They are wrong. In reality, the more people who can ask questions, the more valuable the system becomes. This is because different people have different perspectives and needs, and the AI system can provide answers that are relevant to each of them. For example, a procurement manager may want to know the lead time for a specific component, while a logistics manager may want to know the inventory levels at a particular warehouse.

By giving more people access to the AI system, organizations can unlock new insights and improve decision-making across the supply chain. This requires a cultural shift, where people are encouraged to ask questions and explore the data, rather than just relying on existing reports and dashboards.

The supply chain data this depends on

The AI system depends on a range of supply chain data, including purchase and delivery history, lead times, inventory by location, and demand signal. This data is used to train the AI models and provide accurate answers to user queries. For example, the system can analyze historical delivery data to predict future lead times, and adjust inventory levels accordingly. It can also analyze demand signals, such as sales data and weather forecasts, to anticipate changes in demand and adjust supply chain operations.

By integrating this data into the AI system, organizations can gain a more complete understanding of their supply chain operations and make better decisions. This requires a robust data infrastructure, where data is collected, stored, and analyzed in a way that supports the AI system.

What it changes about lead time variance and inventory risk

The AI system can help organizations to better understand and manage lead time variance, which is often hidden behind a stable average. By analyzing historical data, the system can identify patterns and trends that affect lead times, and provide predictions for future deliveries. This allows organizations to adjust their inventory levels and supply chain operations accordingly, reducing the risk of stockouts or overstocking.

The system can also help organizations to identify and manage inventory risk, by analyzing the likelihood of stockouts or overstocking at different locations. This allows organizations to adjust their inventory levels and supply chain operations to minimize risk and maximize efficiency. For example, a company may hold inventory against the wrong risk, such as holding too much inventory in a warehouse that is prone to stockouts, while not holding enough in a warehouse that is subject to high demand.

Measured with lead time variance, fill rate, and inventory turns

The effectiveness of the AI system can be measured using a range of key performance indicators (KPIs), including lead time variance, fill rate, and inventory turns. These KPIs provide a clear understanding of how the system is impacting supply chain operations, and allow organizations to make adjustments as needed. For example, a company may use the AI system to reduce lead time variance, and measure the impact on fill rate and inventory turns.

By using these KPIs, organizations can demonstrate the value of the AI system and make the case for further investment. This requires a clear understanding of the supply chain operations and the data that underlies them, as well as a willingness to experiment and adjust the system as needed. To learn more about how to implement an AI system for supply chain management, visit our operations page.

What this does not do

The AI system is not a replacement for human judgment and expertise. It is a tool that provides insights and answers to questions, but it is up to the user to interpret and act on that information. The system is also not a one-time solution, but rather an ongoing process that requires continuous monitoring and adjustment. For example, changes in the supply chain or market conditions may require the system to be retrained or updated.

Additionally, the AI system is not suitable for all organizations or supply chain operations. It requires a certain level of data maturity and infrastructure, as well as a willingness to invest in the technology and training. To discuss whether an AI system is right for your organization, contact us at contact-us.