Express Computer recently conducted an insightful interview with Shankar Rao, CIO & CDO of Bosch India, shedding light on the company’s recent digital initiatives and its ongoing digital transformation journey. Rao elaborated on Bosch’s unique approach to fostering a digital culture, emphasising comprehensive digital awareness training through programs like Digital Fluency. He discussed the integration of AI and ML into daily operations, highlighting challenges and initiatives to increase utilisation. Rao also touched upon the traction for GenAI in the industry and outlined future digital initiatives focusing on AI deployment, digital twin technology, and streamlining administrative processes in partnership with government initiatives.
Please talk about some of your recent digital initiatives and your digital transformation journey thus far.
One of Bosch’s distinctive approaches is our emphasis on fostering a digital culture. By culture, I mean a deep understanding of the digital realm and its potential impact across various functions and locations within the organisation. This appreciation is crucial because it enables individuals to recognise how digital technologies can revolutionise their workflows, making them more efficient and effective. To instil this awareness, we initiated a program called Digital Fluency, which involved comprehensive digital awareness training for all Bosch employees in India, excluding BGHW, our software unit, encompassing all manufacturing entities.
During the Digital Fluency program, employees gained insights into available technologies and witnessed real-life applications across industries, understanding the art of the possible. Those displaying particular interest and passion advanced to more specialised training sessions focused on areas like automation and analytics. We identified these individuals as digital pioneers and provided them with advanced, in-depth training sessions.
We have established Centers of Excellence (CoEs) within our organisation, each focusing on distinct areas such as automation, analytics, and next-gen technologies. These CoEs serve as hubs of expertise, guiding and supporting digital initiatives across various functions. For instance, our automation CoE oversees the training of digital pioneers, now referred to as citizen developers, who actively contribute to identifying and addressing challenges through digital solutions.
Our approach involves collaboration between central IT teams, experts, and citizen developers. We provide architectural guidelines and guardrails to ensure scalability and sustainability while empowering citizen developers to contribute to solution development. This hub-and-spoke model enables rapid innovation and implementation, as demonstrated by initiatives like Speedathon, where cross-functional teams collaborate intensively to solve specific challenges within short time frames.
We prioritise leveraging existing solutions before developing new ones, ensuring efficiency and maximising resources. Despite the inherent complexities of digital projects, we adopt an agile sprint approach, allowing for experimentation and exploration of the most suitable technologies.
Through initiatives like Digital Fluency, CoEs, and ongoing competency development programs, we cultivate a digital culture where employees are not only aware of digital trends but also actively participate in shaping technological solutions. This culture shift fosters healthy conversations between IT and business units, leading to collaborative efforts in leveraging technology to address business challenges.
While challenges such as cost considerations and ROI assessments persist, the evolving dialogue between IT and business units reflects a positive shift towards embracing technology-driven solutions. Ultimately, our success in digital transformation stems from fostering awareness, encouraging collaboration, and aligning technology initiatives with business objectives.
How do you integrate AI and ML into your daily operations, considering their relevance in the industry?
We’ve ventured into numerous AI use cases, but the perennial challenge in AI has been the availability and quality of data. Before Gen E emerged and the concept of synthetic data gained traction, data quantity and quality posed significant hurdles. Models can only generate insights based on the data they’re fed. While we’ve launched a considerable number of AI and ML initiatives, the effective utilisation rate remains relatively low. Out of, say, 100 projects where AI/ML was implemented, I’d estimate that only around 20 percent or even fewer actually deliver the desired outcomes aligned with business objectives.
Now, with the advent of GenAI and the concept of synthetic data, there’s a fascinating shift in perspective. Even in the absence of raw data, we can now generate data autonomously. Another crucial aspect is the comparison we’re initiating with business processes over a span of two to three months. It’s akin to running parallel systems: one employing traditional manual methods, and the other leveraging AI/ML-generated data. In these dialogues with our teams, we emphasise why the data generated by the model holds superiority. It considers additional parameters that manual processes often overlook due to inherent limitations. By feeding the model with numerous parameters, it enables a more comprehensive analysis. It’s a continuous dialogue aimed at showcasing the model’s outputs and why they excel, considering the multitude of parameters and scenarios it can accommodate. This ongoing process necessitates a strategic blend of demonstration and explanation, ensuring stakeholders grasp the superiority and accuracy of the model-driven approach compared to traditional methods.
Do you plan to increase the utilisation of AI within the organisation?
We have several initiatives underway, and while there are many projects in the pipeline, the ultimate measure of success lies in their adoption. I can provide you with a sleek application designed to streamline your tasks, but if it sits unused, it falls short of success in my book. The quantity of applications is one thing, but the real metric is how extensively they’re utilised. The goal of AI implementation should always be active usage, and we have tangible use cases across various domains, including maintenance and quality assurance.
Over the past four years, my team has delved deeply into exploring AI applications, particularly in the logistics sector. Considering that we don’t manage our own transportation but rely on over 20 to 30 logistics service providers (LSPs), optimising our logistics operations is crucial. By leveraging data analytics and algorithms, my team has identified opportunities to enhance efficiency, such as recommending LSPs that could potentially reduce costs by 50 paisa per kilometre.
However, the effectiveness of these initiatives hinges on their adoption by stakeholders within the business. Success, in my view, is not just about deploying solutions but ensuring that they are actively utilised by people. It’s common for users to engage with new tools at the outset, but sustaining that usage over time is where the challenge lies. This is where our IT business partners play a vital role, continuously monitoring and supporting the implementation to ensure that the benefits are realised in the long term.
Why do you think there might be a possibility that some entities are not utilising AI, despite its potential to simplify tasks?
The challenge lies in the ever-changing nature of business dynamics. When scenarios shift, it’s crucial for the team to relay updates promptly. If they notice even a minor deviation where the model isn’t delivering the expected results, they halt operations. However, oftentimes, this discrepancy arises from unnoticed changes occurring in the business environment. Hence, it becomes imperative to keep the AI team informed, leveraging the knowledge garnered from business insights. This enables them to fine-tune the model to align with the current landscape.
It’s a continuous journey, an ongoing evolution. I’m certain many face similar hurdles, but our advantage lies in having team members stationed at every plant, within every function, facilitating seamless communication. It’s a perpetual process. While implementing may not pose a challenge, ensuring widespread adoption remains paramount. Hence, my focus is steadfastly directed towards fostering the adoption.
Is it designed for customer-facing interactions or for internal communication purposes?
As of now, it’s internal, primarily within HR and R&D. R&D, in particular, is where we’re delving deeper. Given the substantial volume of data processed within R&D, our aim is to streamline information flow across its five distinct functions. Often, one department may lack visibility into the information available to another, leading to redundant efforts. To address this, we’re consolidating resources to establish a centralised knowledge base accessible to all R&D divisions. This repository will efficiently handle both structured and unstructured data, delivering tailored insights along with the source references.
While R&D is progressing well into this advanced stage, HR initiatives are slightly behind due to competing priorities. However, we’re actively addressing these challenges. Notably, our CEO is keen on harnessing GenAI capabilities. In an upcoming meeting, he plans to explore leveraging external data alongside our internal intelligence for sales and marketing strategies. The objective is to integrate market intelligence with our data to forecast sales trends accurately for specific segments. With the CEO already envisioning the use case and endorsing GenAI as the solution, I’ll be meeting with him and our joint managing director next week to further strategise on this front.
How do you see the traction for Gen AI in the industry?
I believe GenAI is a technology that’s here to stay. It’s not just a passing trend; rather, it has the potential to transform various functions across the organisation. Take HR, for example. Many of our HR business partners spend a significant amount of time answering repetitive queries that could easily be handled by an intelligent system. Similarly, in finance, the role of the CFO has evolved beyond simply reporting past financials. Now, it’s about analysing trends, considering external factors, and engaging in scenario planning. With GenAI, this process becomes more streamlined as it can collect and analyse vast amounts of data, both internal and external, to make accurate predictions.
In manufacturing, where challenges like product returns are prevalent, GenAI can play a crucial role in optimising processes. Whether it’s ensuring customers understand product usage or identifying areas for improvement in manufacturing workflows, GenAI has the potential to make a significant difference.
As we move forward, we’re actively exploring the use of GenAI across all functions. Our teams are evaluating various platforms and technologies to determine the best fit for our needs. The goal is to leverage GenAI to enhance efficiency and effectiveness across the organisation, including data storage and management.
In terms of storage, are you on-prem or on cloud?
SAP, our largest system, primarily resides in our private cloud. While we maintain an arrangement with SAP, the rest of our infrastructure remains predominantly on-premises. However, we’re gradually transitioning towards cloud solutions. One of our flagship products, responsible for outbound logistics tracking, was initially on-premise, but we’re in the process of migrating it to Azure. Additionally, most of our upcoming applications are planned as Software-as-a-Service (SaaS) solutions, leveraging cloud-based platforms. This strategic shift aims to minimise infrastructure costs and reduce the need for extensive in-house maintenance teams, given the inherent advantages of cloud computing.
Adopting a “cloud-first, mobile-first” approach is paramount for us. Enabling mobility is strategic, exemplified by empowering maintenance personnel with SAP Fiori applications accessible via iPads and iPhones. They can efficiently carry out various SAP operations on their mobile devices, enhancing operational agility.
In the realm of digital transformation, we focus on both internal and external aspects. Externally, our aim is to establish seamless digital platforms for engaging with customers, partners, and suppliers. This includes streamlining information exchange with customers, from initial specifications to order status and complaint management, facilitated through a unified customer experience platform. Similarly, fostering transparency and efficiency in supplier relationships is crucial, ensuring real-time visibility into inventory, order status, and supplier performance metrics. Furthermore, leveraging government-provided APIs enables seamless integration with our SAP system, facilitating efficient data exchange with regulatory agencies.
Internally, our focus is on digitising every touchpoint of the employee experience. From flexi-seating arrangements to travel bookings, our employees interact with digital platforms for various corporate functions. We strive to automate and enhance the employee lifecycle, from applicant tracking to onboarding processes, ensuring a seamless digital experience throughout their tenure.
Our digital initiatives span both internal operations and external engagements, with a focus on leveraging technology to optimise efficiency, enhance transparency, and deliver superior experiences across all facets of our business.
How can a balance be struck between utilising AI for risk assessment, and cybersecurity, while also mitigating the potential risks associated with AI itself?
The first part underscores the potential of AI to enhance various aspects, which is widely recognised. However, equally important is acknowledging its capacity to open floodgates, potentially compromising systems or introducing biases that can have severe implications. Striking a balance is imperative, especially from a security standpoint. At Bosch, we adhere to what we call CD, or Central Directives, which provide clear mandates when leveraging AI in any application. These directives encompass a comprehensive checklist of 156 items that must be diligently followed, ensuring that we don’t merely replicate open-source or pre-existing models without scrutiny.
Fortunately, we operate within a robust framework managed centrally, leaving no room for deployment without proper approval. Oversight is provided by a dedicated team, the Data Security Officer (DSO) team, which operates independently from other departments, including my own. Even for seemingly routine tasks like releasing a SaaS application, adherence to CD guidelines is non-negotiable. While the DSO team assesses risks, the onus falls on us to furnish comprehensive documentation detailing our compliance with directives. They function primarily as a governance body, approving based on the evidence provided. While they set the directives, the implementation and documentation remain our responsibility.
Do you have any plans for digital initiatives in the next six to 12 months? Any specific technology or idea you intend to focus on?
GenAI is a key area of focus for us, although we may have been a bit slow to start compared to some other organisations. Currently, we’re still in the proof-of-concept (POC) phase and haven’t deployed anything into production yet. However, by the end of this year, I aim to see tangible progress with something deployed, adopted, and making a noticeable difference. It’s all about creating a clear before-and-after picture.
Another area we’re keen on exploring is the use of digital twin technology in manufacturing for specific processes. Digital twin presents exciting possibilities for optimisation and efficiency improvements.
Also, despite advancements, there are still manual and intensive processes, especially in dealings with government entities like taxation. We’re actively working on streamlining these interactions, leveraging the progressive steps taken by the Government of India to digitise corporate-government interactions. Our goal is to introduce seamless automation across all aspects, whether it’s importing/exporting or obtaining necessary codes, reducing the time spent on administrative tasks and reallocating resources more effectively.
So, moving forward, our focus remains on progressing in these three areas, leveraging emerging technologies and government initiatives to drive efficiency and innovation within our organisation.