• Written by Azhar Aliyar, Director

A supply chain manager checks the purchase order table on a Monday morning, looking for any issues that may have arisen over the weekend, and wonders how AsscherAi can help her get a better view of the supply chain without duplicating the data already in the enterprise resource planning system.

Read paths and write paths

When integrating AI systems with the systems of record for supply chain, it is crucial to consider the read paths and write paths. The read paths are relatively straightforward, as they only involve retrieving data from the systems of record. However, the write paths require more care, as they involve updating the systems of record with new data. If not done correctly, this can lead to inconsistencies and errors, causing the AI system to become a second source of truth.

This can happen when the AI system is not properly synchronised with the systems of record, resulting in outdated or incorrect data being used for decision-making. To avoid this, it is essential to carefully design the write paths to ensure that the AI system is updating the systems of record correctly and in real-time.

Reconciling definitions

Different systems of record may have different definitions for the same data element, which can cause inconsistencies when integrating AI systems. For example, the definition of "lead time" may vary between the enterprise resource planning system and the transportation management system. To reconcile these differences, it is necessary to create a common data model that defines each data element in a consistent manner.

This requires a thorough understanding of the data elements and how they are used in each system. By creating a common data model, the AI system can ensure that the data is consistent and accurate, and that the definitions are reconciled across all systems.

Supply chain data

The integration of AI systems with the systems of record for supply chain depends on several key data elements, including purchase and delivery history, lead times, inventory by location, and demand signal. This data is used to train the AI models and provide insights into the supply chain. The purchase and delivery history provides information on the performance of suppliers and the reliability of the supply chain.

The lead times and inventory by location data are used to optimise inventory levels and reduce stockouts. The demand signal data is used to forecast demand and adjust the supply chain accordingly. By integrating these data elements, the AI system can provide a comprehensive view of the supply chain and help the supply chain manager make informed decisions.

For example, a shift supervisor can use AsscherAi to check the maintenance log and ensure that the equipment is properly maintained, which can help reduce lead times and improve overall supply chain efficiency, as described in operations management.

What it changes

The integration of AI systems with the systems of record for supply chain can change the way supply chain managers view lead time variance and inventory risk. Traditionally, lead time variance has been hidden behind a stable average, making it difficult to identify potential issues. However, with the AI system, supply chain managers can see the actual lead time variance and take steps to mitigate it.

Similarly, inventory risk has often been held against the wrong risk, resulting in overstocking or understocking. The AI system can help identify the correct risk and optimise inventory levels accordingly. By providing a more accurate view of the supply chain, the AI system can help supply chain managers make better decisions and improve overall supply chain efficiency.

Measuring success

The success of the integration of AI systems with the systems of record for supply chain can be measured using several key performance indicators, including lead time variance, fill rate, and inventory turns. These metrics provide a comprehensive view of the supply chain and can help identify areas for improvement.

By tracking these metrics, supply chain managers can see the impact of the AI system on the supply chain and make adjustments as needed. For example, if the lead time variance is increasing, the supply chain manager can investigate the cause and take steps to mitigate it. If you need help with implementing AsscherAi, you can contact us at contact us.

Limitations

The integration of AI systems with the systems of record for supply chain is not a solution for all supply chain management issues. It is not a replacement for human judgment and expertise, and it should not be relied upon as the sole decision-making tool. Additionally, the AI system requires high-quality data to function effectively, and if the data is inaccurate or incomplete, the AI system will not provide accurate insights.

Most people assume that integrating AI systems with the systems of record for supply chain is a straightforward process, but this is not the case. It requires careful planning, design, and implementation to ensure that the AI system is properly integrated and providing accurate insights. Many people are wrong to assume that the AI system can simply be plugged in and will start providing benefits immediately.