Learn, explore and unleash your inner chef.

What type of processes can be automated in order to improve customer facing supply chain/order fulfilment??

In the world of supply chain, we face multiple challenges towards implementation of a demand driven “Material Requirements Planning” (MRP) system. Mitigation of the challenges is key to a successful end-to-end order fulfilment solution for customers. These challenges can fall into one of the below categories:

1. Strategic solution design: This is the brain behind any stable process deployment. We always start from the customers’ requirements and subsequently move backwards to define the requisite supply chain solution. This ensures that we have the optimal stock across the supply chain to ensure continual availability of material with the customer.
2. Deployment of the designed solution: This is where the end-to-end solution is implemented. Based on the complexities of the design, this deployment will have a series of routine/manual and non-routine strategic tasks
3. Continual improvement of the deployed solution: Once deployed, the key performance indicators are continually measured to ensure availability of stock and consequent customer satisfaction. This requires thorough analysis of the “as-is” process to derive smart insights for process improvements.

The challenges are further amplified in large organisations like ours, where geographically separated installations are not often integrated within the core “Enterprise Resource Planning” (ERP) platform. Such systems are often required to communicate with each other to capture demand, raise purchase orders, and ensure smooth material flow.

Automation plays a key role in mitigation of the above challenges. Parts of solution deployment (2) and continual improvement (3) can readily be automated to achieve through-chain efficiencies. These would primarily be the areas where manual intervention is required to progress an order or large amounts of data is available for the system to process and drive data-backed solution improvements. For example, if the Stock Keeping Units (SKUs) do not change, the same order needs to be punched into an upstream system on a regular basis to replenish the stock at a downstream unit that would subsequently flow to the customers. Automating the process around the same for non-integrated systems for all customers (extracting all required order entry fields, identifying first time/repeated orders and taking subsequent automated next steps) could lead to saving of multiple man-hours and greater efficiencies. Similarly, once the system is deployed, an analysis of the orders can reveal variances in “Reliable Replenishment Time” (RRT), “Material Dwell Time” (MDT) ahead of units and the consequent bottlenecks. These are smart insights that can subsequently be used to de-constrain the system, thus managing the service and stock better.

How are you currently deploying RPA and AI solutions?

 We are in the process of deploying Robotic Process Automation (RPA) across multiple stages in the supply chain to reduce dependencies on manual processes. Tata Steel Europe has a global presence with multiple downstream distribution units across UK and Mainland Europe. Such distribution units often have local MRP systems that are not integrated into the upstream supplying hub. Connecting such non-integrated systems that do not share a common ERP platform currently require multiple manual interventions. As explained above, we are in the process of deploying RPA solutions in this area to automate order entry from downstream units and subsequent system integration to ensure full stock visibility.

Artificial Intelligence (AI) solutions are often cognitive. We have tools that are capable of analysing big data arising out of the order fulfilment process. Once analysed, data is translated to actionable intelligence. We are in the process of exploring such a system to accurately predict “Planned Exit Date” (PED) – the date when we expect orders to be ready keeping in view the real-time system constraints. Having the infrastructure for big data analytics to derive actionable insights is a key stepping stone towards building a cognitive AI solution. In future, we could anticipate a situation where the actionable insights feed back into the systems. This will ensure that systems have a cognitive understanding to highlight key risks/variances and advise smart solutions that can be deployed to enhance customer satisfaction. Such risks might not just be internal to our system or plant, but could also arise of key external interference like port strikes, river flooding and snowfall. Incorporating such risk mitigations into our system and subsequent enablement of empowered decisions will be a game changer in the industry. 

 Which benefits have you already realised from the RPA/ AI input?

We have been able to save multiple man-hours of work due to RPA/AI intervention. From an employee engagement perspective, our goal has been to re-route our workforce into satisfying value-added activities. RPA/AI is helping us focus on the same and prioritise on the non-mundane activities that boosts employee morale. This also fosters the innovation spirit in people – something that we gauge through internal employee surveys and is also visible to our customers through enhanced material availability/service offering through cross-functional collaboration amongst stakeholders. 

What would you like to achieve by attending the Automation, Robotics and AI for O2C conference? 

I consider the conference to be a melting pot of supply chain industry experts who are involved in transformations through RPA and AI in their respective companies. The world of RPA and AI is big – we have numerous opportunities to explore. Missing out on key advancements would be synonymous to being left behind at the risk of eventually perishing. This conference presents a great opportunity for the audience to come together, network, share, and learn. I see this conference as a platform where the major supply chain challenges across industries can be discussed, along with ideas around RPA/AI best practices that have helped/can be helpful in mitigating these challenges.