Embracing the Vital Role of Emerging Technologies in Every Organization
Published on : Wednesday 03-01-2024
Shatam Bhattacharyya, Principal Business Consultant, AI Transformation Consulting.
What are the key aspects to consider for a typical company that wants to switch over from a traditional manufacturing process to an automated environment?
Automation will drastically alter the way companies work, and this potential makes it a top priority for the business leaders. During many interviews, leading CEOs from traditional manufacturing companies have identified AI and emerging technologies as a strategic driver in their journey of decarbonisation, sustainable leadership and improving customer experience. Shell is planning to use AI to optimise business efficiencies and profits, over time reducing carbon footprints. “AI has become a very core part of our overall digital transformation journey,” says Shell’s chief AI guru Dan Jeavons. Rio Tinto Chief Executive Jakob Stausholm said the company had committed to increasing research spending to develop technologies enabling its customers to decarbonise. General Motors CEO Mary Barra mentioned on ‘The Claman Countdown’ – “Having an assistant and really being able to use a voice that is clear enough that you can ask questions and get answers, I think that's what artificial intelligence will enable us to do”.
However, traditional manufacturing companies (such as in energy, mining, utilities and manufacturing) struggle in setting bold aspirations, developing robust business cases, leveraging cross-functional capabilities, and devising effective deployment approaches to reap maximum value from automation initiatives. Building these capabilities could help those companies realise benefits at scale, improve customer and employee experience, and build a long-term competitive advantage.
Traditional manufacturing industry’s distinctive nature requires a tailored approach to achieve ambitious business goals and ensure sustainable process and cultural changes. First, the fragmented data landscape and complex legacy infrastructure has posed challenges for heavy manufacturing companies to adopt emerging technologies.
Second, core manufacturing industries operate in a more risk-averse culture compared to other industries, sometimes contributing to distrust in the deep tech such as Artificial Intelligence, Virtual Reality, Augmented Reality, Blockchain, etc. Line workers often prefer a human touch over digital and AI tools, largely because of misconception about changing effective manual processes and automating high-risk processes that affect production or safety. Finally, the labour-intensive operations make it difficult to justify short-term business cases for automation initiatives. Globally, unionised workforce makes reskilling roles around technology and reassigning a significant share of the workforce to higher-value activities challenging. Despite these hurdles, heavy manufacturing industries have pioneered the adoption of industrial automation technology over the last three decades for sustainable operation. They are identifying exciting use cases in automation and analytics, and some have managed to transform at scale.
For successful digital and AI transformations, heavy manufacturing industry leading players need to take five critical steps to address industry’s discrete challenges.
1. Set an aspirational goal for digital and AI transformation and support it with strong executive sponsorship
Setting the tone for successful digital and AI transformation requires the leadership team to weave an ambitious vision into overall enterprise strategy. The vision should be devised in terms of measurable improvement metrics that would add value to the organisation. It should touch upon setting up a robust partner ecosystem to establish culture of digital and AI product portfolio innovation, identifying correct organisational value levers, mitigating enterprise risks, and transforming customer experience.
2. Define customer and employee experience transformation as the key driver for digital and AI transformation
Traditional manufacturing companies need to reimagine their business processes to improve customer and employee experience. The only way to capture the benefit from digital and AI transformation is to redesign the end-to-end process, redefine roles and responsibilities, and reassign people to higher-value activities. Indeed, the key to comprehensive digital and AI transformation is to fundamentally recalibrate employee’s focus on the activities that require human decision-making and problem-solving. Leading manufacturing companies need to increasingly respond to intensifying global competition by putting customer experience and centricity at the heart of digital and AI transformation strategy.
3. Define robust operating model to help employees in adopting to an evolving organisation culture
Technology-enabled transformations often demand cultural changes and reskilling of the workforces. Executive must model these practices and values and clearly articulate the necessity and value of refocusing work. This transformed culture needs to be communicated and reinforced through various methods, shaping employee perceptions, behaviours and understanding.
4. Incorporate a portfolio of technologies with a rugged technology and data strategy
Thanks to easy-to-deploy digital and AI tools, it is now relatively low risk for industrials to incubate more technology solutions and amplify the impact of emerging technologies. However, setting an organisation-wide technology strategy is essential before embarking on the automation and AI journey. Technology strategy should cover guidelines on designing and implementing security and privacy guidelines. Organisations will need to mine new talent pools-not only for technology architects, data scientists, etc., to build the foundation of the technology stack but also to achieve at-scale transformation with in-house capabilities that can weather rapid technological changes and enable agility in day-to-day operation.
5. Repurpose technology implementations as business transformations
Many industry leaders mistake automation, AI transformation or digital transformations as a technology transformation when the goal should be defined in terms of business value realisation. Certainly, technology transformations should take place in collaboration with the IT/data/AI team – but organisations should first validate the technology capability with proof of concept before quickly pivoting to building solutions that yield business value. Proof of concept can establish technology capability within an organisation. In parallel, organisations should create a comprehensive business case and roadmap for business functions to generate investments required to launch and scale technology transformation programs.
Companies can extract value from digital/automation/AI transformation by setting higher aspirations at the outset, defining customer and employee experience transformation as the key driver, defining robust operating model, devising rugged technology strategy and repurposing technology implementation as business transformation. Managed properly, digital/automation/AI transformations can produce significant business impact and provide a strategic impact.
How scalable are automation and digitalisation solutions for different sizes of manufacturing facilities?
Every organisation would need to reinvent its operating model to win in the digital and AI world. How to deliver services and build products profitably and cost-effectively has become a defining challenge for businesses today. One of the key reasons is that executives face an increasingly complex landscape of technologies, methodologies, and both regulatory and compliance pressure to ensure that new processes are standardised and traceable. Every organisation has realised the importance of adopting emerging technologies.
“The business world will be dominated by those capable of mastering the interfaces between advanced digitalisation and unique specific industrial and scientific knowledge, bringing massive differentiation in cost and market opportunities in the years to come.”- Greg Ludkovsky, Vice President, ArcelorMittal Global R&D.
Business leaders must “accept the new reality”, Holcim CEO Jan Jenisch says, arguing that organisations have no choice but to embrace megatrends such as digitalisation, diversity, and sustainability to earn their place in the future.
However, traditional manufacturing companies are facing different challenges in the journey of technology adoption.
First, organisations do not devise an overall business transformation goal before embarking on a technology adoption journey. When technology solutions are implemented in silos, marginal value is realised. However, companies get stuck in that phase and fail to realise the full potential. Second, multiple business functions start to deploy use cases of automation and AI on their own without any central guideline on technology strategy or organisation prioritisation. Ultimately, organisations end up spending 4x-5x compared to what they would have spent with a central vision and strategy. Due to enormous spending, the automation/AI transformation programs are scrapped or stuck. Finally, companies do not plan on setting up a workforce for scaling transformation capabilities. Due to scarcity of resources and high salary cost, organisations are not able to build a team for operationalisation of automation/AI solutions. This prevents the scaling of automation.AI solution for realising full value.
For successful automation and AI transformations, core manufacturing industry leaders need to build a Centre of Excellence (CoE). This will help them in scaling for different sizes of manufacturing facilities. The CoE should have following guiding principles.
Set a vision
Setting the tone for successful digital and AI transformation requires the leadership team to weave an ambitious vision into execution strategy for the CoE. The vision should be devised in terms of measurable improvement metrics that would add value to the organisation. It should touch upon setting up a robust partner ecosystem to establish culture of digital and AI product portfolio innovation, identifying correct organisational value levers, mitigating enterprise risks, and transforming customer experience.
Measure the business value
Many industry leaders mistake automation, AI transformation or digital transformations as a technology transformation when the goal should be defined in terms of business value realisation. Certainly, technology transformations should take place in collaboration with IT/data/AI team- but organisations should first validate the technology capability with proof of concept before quickly pivoting to building solutions that yield business value. Proof of concept can establish technology capability within an organisation. In parallel, organisations should create a comprehensive business case and roadmap for business functions to generate investments required to launch and scale technology transformation programs.
Setup policies and procedures
Successful companies need to constantly rethink how to bring together the right combination of processes and policies to deliver automation and AI initiatives at scale. That means defining standard processes for managing demands across different business functions and setting the priorities right at organisational level rather than business function level. The demand management process should be designed based on maximum return to the shareholders. The process should be flexible enough also to empower teams for owning products, services as well as running experiments in the proof-of-concept stage. For handling large technology portfolios, a robust software lifecycle management process needs to be designed. Working AI committee needs to take care of defining these processes and policies and managing them throughout the execution lifecycle.
Flexible and modular architecture, infrastructure and software delivery
Scalable technology architecture is a core element of any next-gen technology CoE. It needs to support much faster and more flexible deployment of products and services. However, companies need to have a clear understanding of how to implement these technologies alongside legacy systems. To address these challenges, leaders need to build modular architecture that supports flexible and reusable technologies. Leading technology teams collaborate with business teams to understand the pace of change. Accordingly, they can architect modular solutions with flexibility of speeding up builds and reducing maintenance. The technology components should be selected based on this guiding principle only. This approach accelerates development and prioritises the use of common components, leading to development efficiency and consistency.
Foster culture of collaboration internally
The CoE should be designed on principles, tools, and associated behaviour that drives a culture of continuous improvement. The teams should be designed with cross-functional skills. It should work in partnership between business and IT (such as IT infrastructure, security, legal, compliance and product development). Sharing of best practices should be promoted and should be part of the performance monitoring system.
Setup partner ecosystem
To thrive in the era of emerging technologies, every organisation must continuously innovate and keep updated with the latest technology developments. From cloud to artificial intelligence, change acceleration and more, an open ecosystem should be built to allow sharing of new technology development and take decisions on building a proof of concept or adoption of new technology at an early stage of development.
Companies can set up a successful CoE by setting a vision, measuring business value, setting up policies and procedures, building a flexible and modular technology operation ecosystem, fostering a culture of collaboration and setting up a partner ecosystem. Managed properly, the CoE can help in scaling up and realising business value across multiple manufacturing facilities.
What are the initial costs associated with implementing factory automation and digitalisation?
The costs associated with implementing factory automation and digitalisation can be categorised in one-time capital expenditure and recurring variable cost. One-time capital expenditure covers procurement of necessary hardware, software, etc. Recurring cost includes incurred software license cost, maintenance cost, manpower cost from the CoE setup to scale and run the operation, etc.
How does the adoption of automation and digitalisation impact the skills required for the workforce?
How companies navigate the technology world to sustain competitive advantage is the ultimate challenge for many CxOs of the core manufacturing industry. To be fair, this challenge has been persistent over the last decade. But the business implications are getting broader day by the day with the advent of emerging technologies like AI, blockchain, etc. Companies realise they need to respond to this challenge, but they are struggling. Yet it’s also a challenge with enormous potential for those who get it right. Industry leaders in other industries like BFSI, Telecom, etc., have outperformed their peers with a deeper integration of technology across end-to-end core business processes.
During this journey, these industry leaders have prioritised workforce reskilling, building the right talent base and improving technology adoption. Currently, heavy industries face few challenges in their adoption journey. First, the core manufacturing companies have traditionally outsourced large technology transformation programs. They have emphasised on building in-house competency on core operations. On the other hand, other industry leaders have leveraged a robust in-house CoE model to embark on a transformation journey with emerging technologies.
Second, heavy industries have looked at IT, digital and AI talent as silos. They have not looked at these talents as an integral part of the business transformation journey. On the other hand, other industry leaders have relied on cross-functional capabilities to progress in the transformation journey. Finally, the field workforce has prioritised their day-to-day activities in most of the cases due to the inherent safety hazards associated with their roles. A fundamental shift in technology adoption needs to happen for most of the field workforce; because their inputs are very essential to bring in a digital-first or AI-first mindset during the transformation journey.
Now let’s have a look at how these industry leaders accomplish this and what steps can core manufacturing industries take to increase adoption of automation and digitisation impact?
Build talent bench
It is not impossible to outsource the way to digital and AI excellence. Being digital and AI-first means having a bench of talent-product owners, experienced designers, cloud engineers, software developers, and so on-working side by side with business colleagues. Digital and AI transformations are first and foremost people transformation. The two actions every executive should take.
Build talent base from scratch
Most organisations hire digital technologists and data scientists now-a-days, but may still face the hard work of reskilling their technology and IT organisation. Large portion of the talent base should be nurtured in house while niche and specialised skills should still be outsourced. Talent pyramid should shift to diamond shape, with more competent technologists and fewer novices. That drastically changes the productivity gain. Companies also should focus on having hands-on-technologists at a considerably higher proportion compared to managerial roles. The hands-on-technologists with special skills may be outsourced.
Be more structured about skillset
Transformed companies develop very granular skill progression grids supported by credentials. Without a precise calibration of skills, it becomes difficult to recognise distinctive skillsets and compensate them accordingly. Skill progression translates into expert-based career tracks with tailored learning and development plans. In short, the whole talent model revolves around fostering excellence in people to build their craft.
These shifts in talent practices are not straightforward, but they are fundamental to becoming rewired with the right talent.
Unlock adoption and scaling
Developing a good technology solution can be difficult and complex. But getting business users to adopt these solutions in their day-to-day activities and then scaling that solution at enterprise level is more challenging. Successful companies have concentrated on three moves.
Focus on adoption from day one
Adoption should never come as an afterthought. It should be part of organisation DNA from day one of development. User adoption often starts with delivering great end-user experience. But companies often underestimate the linkages to business models that need to be changed to secure adoption. It may revolve around redesigning the current process. It may be about educating the end user on new ways of working assisting them in understanding the benefits. It revolves around showing them the improved results with help of the tool. End-to-end system approach with a dedicated focus on people slide is what differentiates the digital and AI leaders.
Scale with assetisation
Replicating the adoption of solutions in different environments, such a network of manufacturing facilities, or in different geographies or markets, customer segments, or organisation groups, can be challenging. Companies often end up spending a lot of time in contextualising the solution to a newer market or newer customer segment or newer organisational group. So, all organisation leaders need to focus on making the most of the solution component as reusable during the development of the solution initially. Critical success factor lies in involving all the key decision makers from different user groups at the early stage of digital or AI solution development, irrespective of execution timeline. Incorporating any large customisation at a later stage increases the execution complexity and cost, resulting in stalling of the solution adoption.
Track the correct KPI
It is essential to track the progress of the digital and AI transformation journey from day one. But the question lies in what to measure and how. Performance tracking mechanism that is poorly designed and that is lacking the correct support tools can crumble under its weight. Rewired companies make the cross-functional pods within the CoE responsible for progress in the digital transformation journey. They link the success to key operational metrics. The progress of pods is reviewed regularly through stage gate reviews.
The ability to capture the full economic potential of digital and AI innovations is a core differentiator between industry laggards and industry leaders. Building this capability is a signature move that differentiates the business leaders from the rest.
The capabilities laid out here are essential in the adoption journey of any digital and AI solution. To borrow Jeff Bezos’s expression to Amazon shareholders about the importance of operating like a digital native: ‘It’s always day one for digital and AI transformation’.
What regulatory considerations should manufacturers keep in mind when implementing automation and digitalisation with respect to safety and security?
It wasn’t too long ago that sophisticated executives could have long, thoughtful discussions on technology strategy without mentioning safety and security.Today, organisations have invested substantial assets and value manifested in digital form, and they are deeply connected to global technology networks. The newer investments in emerging technologies bring more exposure to changing regulations. Different government institutional bodies are also bringing new regulations regularly to tackle the threats emerging from new technologies. For example, deepfake incidents have created a national crisis to full-blown public distrust in the information spread. Here is what organisations should do in the context of changing regulations.
A holistic approach to evolving the regulatory landscape proceeds from an accurate overview of the risk landscape. The goal is to empower themselves in this changing scenario, achieving a right balance between effective resilience and efficient operations. The holistic approach should adhere to the following guidelines.
Identify necessary regulations related to data protection and security
In the new era of AI and digital transformation initiatives, the new regulations from data protection would have large implications on the transformation programs. Organisations should pre-empt these changes with active participation in the regional regulatory bodies. They should opt for service providers who have adequate expertise and knowledge of evolving regulations in this space. Privacy laws around the world mandate how organisations should use data, while customer expectations set normative standards. Ignorance to these laws and norms can result in significant liability, as well as harm to consumers. Many legal frameworks mandate data security provisions to avoid vulnerabilities such as model extinction and data poisoning.
Comply with emerging AI acts and laws globally
Any new AI transformation initiative should have a checkpoint against the latest AI laws and acts of all the countries. This should be checked regularly by a cross-functional team from the working AI committee.
Form a risk governance structure
Every organisation should have a formalised risk control and governance in place. Working with top executives and drawing on internal and external resources, the chief risk and information security officer should create a list of critical assets, known risks and potential risks. In conjunction with this effort, organisations should have the appetite of prioritising the identified risks. An assessment should be made periodically to identify existing risks and vulnerabilities. Once the risks and threats have been identified, they need to be classified in terms of different risk categories like regulatory risk, reputational risk, brand risk, operational risk, etc. Based on this assessment, the mitigation plans need to be prioritised. Once the risk has been identified and prioritised, an owner should be assigned and a roadmap should be defined. The initiatives should be evaluated based on their effectiveness in reducing the probability of a risk event occurring and impact of an event that does occur. Should the residual risk levels exceed considered organisational limits, an additional mitigation plan needs to be defined. Among the most important instruments for fostering discipline throughout the organisation are scheduled status updates to senior executives on key risks identified and their mitigation strategy.
Industries, sectors, and regions around the world have varying standards and laws regarding the various risks mentioned above. Therefore it’s important to be aware of applicable laws and regulations based on where and how digital applications or AI models will be deployed.
How can existing machinery and systems be integrated into a digitalised manufacturing environment, and the challenges in the integration process?
Amid rising competition, commoditisation, and price pressure, digital and AI transformation are crucial for machinery and production plant manufacturers.
Traditionally for good reasons, leading machinery and manufacturing companies have focused on selling proprietary hardware and machinery and based on success and business model on the quality of the product. However, the following aspects may need to be relooked by them.
Commoditisation of hardware
Hardware – in the sense of machinery and production plants – has largely become commoditised. Time has come to start substituting specialist knowledge with software in certain parts of machine manufacturing. That is all required to produce today’s machinery. Today’s new technologies, such as additive manufacturing, enable machine manufacturers to produce more complex designs, which traditional manufacturing techniques did not support, and fulfill customer demand at a lower cost. Hence hardware should be commoditised to assist in accelerated adoption of new technologies.
Semantic shift in technology stack value pools
Market trends are causing a fundamental shift in value concentration within digital and AI technology stack from the hardware layer to hardware-software-service offers. These are now becoming less dependent on hardware, and becoming more reliant on the new software and service providers in the market. These players are light on domain knowledge. But they possess superior understanding of emerging technologies and analytics/AI capabilities. They are using their competencies in gathering data and offering innovative services in the space of AI, analytics, AR, VR, etc. However, the complexity lies here on the integration of data. Most of the traditional manufacturing facilities machineries are not ready to expose the data in required format for the new age technology players. Hence, a massive capital expenditure is incurred in transforming the data to required format and structure.
To address both the challenges, it is essential to enter the journey of data transformation. Data should be thoroughly sorted and organised for easy consumption and reuse. They should focus on building reusable building blocks-data products. A data product delivers a high-quality, ready-to-use data set in a way that people and applications across the organisation can easily access and consume. Companies can prioritise building data products that have broader applications, generate maximum business value, and are unique in nature. They should focus on building a system of pipes that deliver the data from where it is stored to where it is used. When implemented well, data architecture fastens a company’s ability to build reusable and high-quality data products and put data to within reach of any team in the organisation. The emergence of new architectural patterns such as data-lake makes it easier for companies to solve both their business intelligence problems and AI needs. Advanced organisations in this journey deploy a central set of guidelines that sets policies and standards along with necessary oversight.
(The views expressed in interviews are personal, not necessarily of the organisations represented.)
Shatam Bhattacharyya, Principal Business Consultant, AI Transformation Consulting is a seasoned business consultant with 11 years of experience in process transformation by leveraging AI.
Shatam has prior experience in the BFSI, FMCG and manufacturing industries spanning across multiple functions like supply chain, manufacturing and IT services. He has helped clients design and drive transformation programs to maximise realised value. He has worked with clients across 30+ countries including 10+ European countries; UK, USA and Australia; 5+ Latin American countries; 8+ Asian countries including India; and 7+ African countries.
Shatam is a resilient, humble, and dedicated individual who has excelled consistently across different roles in his career. He believes in creating a high performing team by developing competencies through a culture of collaboration and mentoring people. He has managed a team of 5+ members while juggling 2+ project teams continuously over the last 2+ years.
Currently, Shatam is part of AI Transformation Consulting within Infosys Consulting and has contributed significantly to opening two new accounts and winning two large deals for IC in India and Middle East. He is the offering owner of the Infosys proprietary offerings “Responsible AI” and has played a pivotal role in conceptualising the offering which focuses on building organisation’s competitive advantage by leveraging AI with responsibility along with the practice head.