Allcargo’s Kapil Mahajan Predicts Convergence of Quantum, Edge, and Hyper-Localisation to Redefine Technological Landscape in 24 Months

In an exclusive interview with Express Computer, Kapil Mahajan, Global Chief Information Officer, Allcargo Logistics Ltd., provides insights into the profound impact of digital transformation on the logistics and shipping industry. The interview delves into Allcargo Group’s pivotal role in shaping the evolution of the sector and its commitment to leveraging cutting-edge technologies.

The logistics industry, traditionally slow in technology adoption, witnessed a significant paradigm shift in recent years. Kapil Mahajan emphasises the industry’s recognition of technology’s pivotal role, particularly amid the COVID-19 pandemic. Allcargo Group stands out by not only following industry trends but by strategically investing in digital interventions to enhance operational efficiency and overall business agility.

How has digital transformation specifically impacted the logistics and shipping industry, and what role has Allcargo Group played in shaping this evolution?

Traditionally, the supply chain and logistics industry, both in India and globally, has not been inherently digital-native, lagging in technology adoption. However, a significant shift occurred in recent years, notably pre-COVID, during the pandemic, and post-COVID. Businesses started recognising the pivotal role of technology and digital interventions in the industry. In India, many organisations embarked on this digital journey, with some already achieving maturity. Globally, beyond a few major players, a widespread trend of substantial investments in digital initiatives emerged, introducing industry-first use cases in the past three to four years. In our specific case, we have dedicated substantial efforts and resources to digital interventions, focusing not merely on following the trend but rather on enhancing operational efficiency and overall business agility.

Given the low-margin nature of our industry, interventions that impact cost reduction, process efficiency, and reduced dependency on manpower hold significant value. Recognising this, we have established a Chief Digital Officer role, leading a team dedicated to digital-first projects. We have not only adopted digital at the surface but also strengthened core systems to align with the digital ecosystem comprehensively. It is crucial to integrate digital transformation at all intersection points, emphasising its impact on the bottom line. At Allcargo, our substantial investments in technology have resulted in the digitalisation of numerous processes.

Digital intervention is an ongoing journey, encompassing AI, machine learning, and more recently, generative AI. While translating these technologies into mainstream business practices requires research, development, and proof of concepts, we are committed to this investment. From a group standpoint, our vision is to infuse digital capabilities across every business process. For instance, our group company, Bathy, is pioneering the development of a cloud-native ERP, a first-of-its-kind in the industry, representing a long-term investment aimed at transforming logistics practices in India.

In what keyways you are leveraging artificial intelligence and machine learning to optimise the logistics operations and enhance overall efficiency?

In the past four years, I have strongly advocated for the idea that every business process will evolve into a cognitive one, a vision that is materialising today. Beyond just a few use cases, mainstream AI is expanding its reach, and it is imperative to democratise AI, making it a competitive tool in the hands of operational users. This goal aligns with ensuring that every process becomes smarter, augmented, and cognitive. An example is the adoption of Computer Vision, a form of AI, to handle unstructured data like videos and images in our logistics operations. For instance, our computer vision model automates the interpretation of delivery documents, enhancing efficiency by eliminating manual data entry.

Another application involves the use of AI in proof of delivery processes, improving the cash flow cycle by enabling real-time verification and immediate invoicing. Fleet optimisation is also enhanced through AI, optimising vehicle placement at cross docks for more efficient loading and unloading. Also, AI and machine learning assist in determining optimal routes for last-mile deliveries based on locational data. To democratise AI further, we have incorporated offline AI capabilities, utilising edge computing applications that work offline, ensuring functionality even in areas with limited connectivity. We have focused on tailor-made AI solutions, deploying cloud-based models globally and offline models within India, catering to specific operational needs. While some use cases remain confidential, these advancements underscore our commitment to leveraging AI and machine learning to enhance operational efficiency and drive innovation within the group.

Given the rise of generative AI, how do you see its potential applications in the logistics domain, and have you explored any specific use cases for generative AI?

Currently, our utilisation of generative AI is in an experimental phase, with a specific focus on improving our HR processes. We recently introduced a generative AI-powered conversational bot designed to handle HR documents. The model is trained on a diverse range of HR policies from our three major entities—Allcargo, ECU, and Gati. When a user poses a question, the model generates responses by extracting relevant information from the documents, utilising generative AI. The primary objective is to enhance the employee experience by enabling the generative AI to promptly address common queries related to leave policies, medical benefits, and other HR-related matters.

This initiative aims to alleviate the workload on HR personnel, allowing them to focus on more complex tasks. Additionally, we are exploring other use cases, such as content creation, leveraging GenAI. However, we are mindful of potential copyright issues and are working with partners like GCP and Amason to train the model on our datasets. While the adoption is gradual and teams are still in the learning phase, the HR bot has shown promising results, laying the foundation for broader applications of GenAI within the organisation.

Do you have any plans to make it accessible for external communication or the stakeholder communication component?

In our ongoing exploration of AI applications, we are strategically considering potential use cases to avoid copyright concerns and maximise the impact of our initiatives. One promising area we are delving into is utilising generative AI models to enhance customer interactions, particularly in addressing concerns raised in email communications. For instance, when a customer expresses dissatisfaction with the delay in receiving information about their shipping consignment or a quote, we aim to leverage generative AI to intelligently generate personalised responses.

By assessing the tone of the customer’s messages and the context of their interactions, the AI model can craft appropriate and empathetic replies to pacify the customer and provide the necessary information. This approach seeks to streamline and automate responses to the first and second levels of communication, enabling our staff on the ground to focus on more complex queries. Ultimately, we anticipate that this initiative will not only improve customer satisfaction but also enhance our overall business prospects by efficiently handling a higher volume of quotation requests and inquiries. As we proceed, rigorous testing and assessment of the model’s performance will guide the decision-making process before full implementation.

Allcargo currently operates in approximately 180 countries worldwide, indicating a significant volume of data handling. Given the growing dependence on data-driven technologies and the implications of the Digital Personal Data Protection (DPDP) Act, 2023, could you please provide details on the measures you have taken to safeguard data security and privacy in your digital initiatives?

Regarding the India DPDP Act, our approach aligns with our global practices to comply with data residency laws specific to each country. Operating in 180 countries, we have implemented standard practices across six global data centres to ensure adherence to respective data residency policies. In India, where the data privacy regulations have been reinforced, we have proactively taken measures to store Indian data within the country. Considering the potential penalty of around 250 crores for unauthorised leakage of sensitive Personally Identifiable Information (PII) data, we recognise the importance of robust data security measures.

While our B2B focus typically involves limited storage of personal financial information, we are diligently working on enhancing data security measures. This involves exploring tools that enable data concurrency and implementing mechanisms to track and monitor access to sensitive data. Evaluating advanced tools that provide insights into data access, including details such as frequency, specific information accessed, and relevant system and IP addresses, is a crucial part of our compliance strategy. Despite the associated costs, we are committed to ensuring compliance with the DPDP Act and investing in the necessary tools for data security and privacy.

In the context of your digital transformation journey, have you encountered or are you currently experiencing any technological challenges? Additionally, do you have identified areas for improvement in terms of technology, and if so, what measures are being considered for enhancement?

When we talk about challenges, a significant aspect revolves around the workforce, particularly acquiring the right talent. In the realm of technology, success heavily relies on the calibre of human resources. Poaching individuals from the mainstream tech industry to join an end-user company poses difficulties, especially when competing for the brightest minds, and the compensation expectations can become extravagant. These challenges persist and are likely to endure. However, from a technology perspective, there is not a notable gap. We recently introduced shared services, incorporating seven verticals, with enterprise architecture being a centralised shared service managing technical referential architectures across the group. This ensures standardisation and best practices.

Additionally, the development, engineering, and testing competency, including automated testing, has been established as a shared function. This structure allows talent to cut across roles, provides more opportunities, and creates a larger team for cross-leverage and reskilling. While the challenge in terms of talent has been somewhat mitigated through this approach, on the technological front, we have adopted a strategy of consolidating technology stacks. Instead of supporting numerous languages, we focus on the latest stack, investing in a resource pool that can work across organisations. For instance, if Salesforce is the chosen CRM platform, we create a centre of excellence and a shared pool working on different applications for various companies, ensuring common underlying architecture and optimised costs. Overall, with the latest tech stack, including AI and machine learning, and a sizable team in data science practices, we do not see technology itself as a major challenge.

Looking ahead, what trends do you anticipate in the intersection of technology and logistics, and how is Allcargo positioning itself to adapt and thrive in the evolving landscape?

My transition from IBM, where I spent over a decade and a half, to the industry was driven by the fascinating realm of Express business, propelling me towards the dynamic field of supply chain. It is an industry teeming with potential for technological advancements, albeit not yet at its zenith. The diverse applications of emerging technologies such as IoT, AR, VR, AI, ML, generative AI, and blockchain within the end-to-end supply chain make it an exciting space to explore. While AI, ML, and generative AI have gained substantial attention, I am particularly enthusiastic about the advent of quantum computing.

With IBM launching quantum computing products, there is a possibility of it becoming mainstream by 2024-2025, potentially reshaping the landscape of computing. The imminent shift towards edge computing is another trend I anticipate, with most computing operations migrating to end-user devices, fostering hyper-localisation and a more user-centric ERP experience. As we progress, I foresee BI becoming an integral part of every business process, seamlessly ingrained for predictive and prescriptive analytics, revolutionising the current analytics landscape. In the next 12 to 24 months, the convergence of quantum computing, edge computing, and hyper-localisation is poised to redefine the technological landscape, ushering in a new era of possibilities.

As a Chief Information Technology Officer or as a tech enthusiast, what is that one technological innovation that you are particularly proud of?

Okay, interesting. I am proud of few. I will share with you one example from my tenure as the head of the analytics practise for a prominent American MNC in the group insurance sector. About a decade ago, we grappled with the challenge of FMLA (Family and Medical Leave Act) leaves being misused within the organisation. Leveraging complex statistical modelling, we harnessed over 200 data points, exploring numerous permutations and combinations. The resultant predictive model enabled us to forecast FMLA trends within the workforce, vital for an insurance customer’s legal obligations.

By sharing this predictive data, manufacturing companies could strategically plan production life cycles, orchestrating shutdowns on days with a high probability of employees taking FMLA leaves. This not only optimised production processes but also translated into significant cost savings, garnering accolades for the project.

Another notable endeavour involved collaborating with Pfizer on a healthcare initiative utilising IBM Watson. Leveraging historical patient data, we developed a predictive model capable of identifying individuals at a higher risk of a heart attack. This innovative project enabled proactive interventions, recommending pre-emptive medications and surgical procedures for those identified as high-risk cases. The impact extended beyond financial considerations, showcasing the transformative potential of predictive analytics in healthcare. These projects exemplify the varied dimensions of analytics, demonstrating its capacity to influence both operational efficiency and, more significantly, enhance lives.