AI-Driven Supply Chains: Building the Future of Intelligent Manufacturing

In an era of rapid digital transformation, Artificial Intelligence is redefining supply chain management, inventory optimisation and enterprise operations. In this conversation, Santanu Boral, Associate Director & Solution Architect, Cognizant Technology Solutions, US, shares his perspectives on leveraging AI, predictive analytics and multi-agentic systems to build agile, resilient and intelligent supply chains. He also highlights the importance of data governance, human expertise and strategic execution in enabling India’s engineering and manufacturing industries to achieve greater efficiency, competitiveness and sustainable growth.- Editor

Q. India’s engineering and manufacturing sectors are undergoing a significant digital transformation. In your assessment, how mature are Indian organizations in adopting digital technologies, and what are the key opportunities and challenges in accelerating this transformation?

India’s manufacturing sector is at a true inflection point, advancing rapidly but unevenly. Large enterprises such as Tata Steel, JSW Steel, Mahindra, and Bajaj have already moved AI and Industry 4.0 from pilots to full-scale deployment, achieving measurable gains in predictive maintenance, computer-vision quality inspection, and energy optimization. Tata Steel Kalinga Nagar’s recognition as India’s first WEF Industry 4.0 Lighthouse shows what world-class digital maturity looks like. Yet hundreds of mid-tier manufacturers and MSMEs — contributing over 35% of India’s manufacturing value added — still operate with fragmented legacy systems and manual processes.

The opportunity is substantial. India’s smart factory market is expected to grow from $7.7 billion in 2025 to nearly $17 billion by 2032. The PLI scheme and the India AI Mission’s ₹10,300 crore investment signal strong policy intent. As manufacturing aims to rise from 17% to 25% of GDP, digital capability becomes essential.

The biggest barrier is not technology but data. Siloed systems, fragmented ERPs, inconsistent quality, and change resistance slow progress. Digital transformation must be treated as business transformation — anchored in clear problems and enabled by cloud ERP, IoT connectivity, and strong data governance.

Q. With over 25 years of experience in supply chain technology, how do you see Supply Chain Management (SCM) solutions evolving to meet the changing needs of engineering and manufacturing enterprises? What capabilities will be critical for building agile, resilient, and responsive supply chains?

Supply chain management has evolved more in the past five years than in the previous two decades. What was once ERP-centric transaction processing has become a real-time, intelligence-driven capability central to competitive strategy. The biggest shift is from retrospective reporting to anticipatory, cloud-enabled, prescriptive decision-making.

Three capabilities now define advanced supply chains. First, end-to-end visibility — sensing disruptions, inventory positions, supplier risks, and demand signals across multi-tier networks through IoT data, collaboration portals, and logistics tracking. Second, AI-powered planning that uses external signals such as weather, economic trends, sentiment, and promotions to generate probabilistic forecasts. Third, resilience architecture — deliberate redundancy, dual sourcing, and scenario modeling so organizations respond to disruptions in days, not weeks.

Finally, multi-agentic AI will reshape SCM. Instead of monolithic models, networks of specialized AI agents collaborate across forecasting, replenishment, and supplier risk, escalating decisions to humans when confidence is low. This delivers both AI speed and human judgment — essential for complex, regulated supply chains.

Q. Inventory optimization is a persistent challenge for manufacturers, with excess inventory tying up capital and shortages disrupting production. How can AI, predictive analytics, and machine learning help organizations improve demand forecasting, optimize inventory, and enhance service levels?

Inventory optimization sits at the intersection of finance, operations, and customer experience, yet remains a persistent inefficiency for most manufacturers. AI-driven inventory optimization is replacing traditional planning models, and the shift is accelerating. The global AI inventory optimization market, valued at $5.9 billion in 2024, is projected to reach $31.9 billion by 2034 — a clear signal of urgency.

Traditional models rely on static reorder points and historical averages, which work only in stable environments. Modern supply chains are anything but stable. AI fundamentally changes the equation. Machine learning continuously learns from demand patterns, promotions, supplier variability, seasonal signals, and external datasets such as macroeconomic indicators and social sentiment. The result is a dynamic, probabilistic demand signal that updates in near real time.

Research shows AI-driven forecasting can improve accuracy by up to 90%, reduce stockouts by 35%, and cut excess inventory by 28%. Manufacturers should embed AI into core planning, starting with high-velocity SKUs and expanding to long-tail and spare parts. The goal is precise inventory — the right stock, in the right place, at the right time, with minimal capital.

Q. AI-powered supply chains depend on accurate, integrated, and timely data. What are the major data-related challenges facing manufacturing organizations, and how can enterprises build the digital foundations needed to translate data into actionable business intelligence?

Data fuels AI, yet most manufacturers face data abundance but insight scarcity. ERP, MES, IoT, and logistics data remain siloed, inconsistent, and poorly governed. Before pursuing AI driven decision intelligence, enterprises must fix fragmented data foundations.

The challenges fall into four categories: Data fragmentation — multiple ERP instances and legacy systems create conflicting records of the same event. Data quality — duplicates, missing fields, incorrect units, and stale master data corrupt AI model training. Data latency — overnight batch updates are insufficient for real time sensing or disruption response. Data governance — without ownership, stewardship, and lineage, even clean data degrades over time.

Building digital foundations requires harmonizing master data, establishing a single source of truth, and modernizing cloud data platforms. A governed, well architected data lake transforms fragmented data into a strategic asset. Analytics and AI must be built with business outcomes as the anchor — every dashboard or model should answer a question planners can act on.

Q. As AI becomes increasingly embedded in enterprise systems and workflows, how will it transform traditional supply chain functions such as procurement, production planning, supplier management, logistics, and order fulfillment? Which applications do you believe will deliver the most immediate and measurable business value?

The embedding of AI across core supply chain functions is the most profound operational shift I’ve witnessed — more transformative than the move from paper-based workflows to ERP. AI expands what is possible through speed, scale, and pattern-recognition capabilities far beyond human teams.

In procurement, AI automates supplier discovery, contract analysis, spend categorization, and risk scoring. NLP models scan thousands of contracts in minutes, while predictive algorithms track supplier financial health and geopolitical exposure, reducing disruption incidents by 20–30% within a year.

In production planning, AI replaces static scheduling with dynamic, constraint-aware optimization that adjusts sequences based on demand shifts, machine availability, and material constraints — especially valuable in regulated industries.

Logistics and fulfillment see immediate gains through real-time route optimization, AI-powered order promising, and warehouse automation using computer vision and autonomous robots.

Among all applications, AI-enabled demand sensing and multi-echelon inventory optimization deliver the fastest ROI, turning AI from a technology upgrade into a strategic business imperative.

Q. The next era of enterprise transformation requires technology, data, governance, and human expertise to work together. How can organisations establish the right balance between AI-driven automation and human judgment, particularly in critical operational and strategic decisions?

The tension between AI automation and human judgment is often framed as a binary choice, but that framing is flawed. The real question is which decisions AI should make, under what conditions, and with what governance structure. My approach begins with decision classification across two axes: frequency and consequence. High-frequency, low-consequence decisions — such as replenishment triggers, carrier selection within approved lanes, or routine PO creation — are ideal for full automation because AI executes them faster, more accurately, and with bounded risk.

High-consequence, low-frequency decisions — supplier strategy shifts, network redesign, major capital commitments — demand human judgment supported by AI-driven scenario modeling, risk quantification, and option generation.

Governance is essential. As highlighted in KPMG’s 2026 Global Tech Report, managing AI agents will soon become a critical workplace skill. Human roles will shift from executing decisions to governing the systems that make them. This requires new competencies in AI literacy, explainability, bias detection, and override management. Trust depends on transparency and a culture where human overrides strengthen the system rather than undermine it.

Q. Despite growing investments in digital technologies, many organisations struggle to scale pilot projects into enterprise-wide solutions. From your experience, what are the key success factors for implementing digital supply chain solutions, and which KPIs should manufacturers track to measure their impact on productivity, cost, resilience, and profitability?

The pilot-to-scale gap is one of the most persistent challenges in digital supply chain transformation. Most pilots succeed in controlled environments but fail to scale because the barriers are organizational, data-related, and governance-driven rather than technical.

From my experience, five factors determine success: First, anchor every initiative to a quantified business outcome — not a technology mandate. Targets such as reducing finished-goods inventory by 15% while maintaining 98.5% service levels drive alignment. Second, prioritize data quality over model sophistication; clean, governed data consistently outperforms advanced algorithms trained on poor inputs. Third, secure active executive sponsorship to break functional silos. Fourth, deploy iteratively — launch a production-ready pilot in a high-impact domain, prove value within months, then scale horizontally. Fifth, treat change management as a core workstream, not an afterthought.

Manufacturers should track KPIs across productivity (OEE, planning cycle time), cost (inventory days, freight cost per unit), resilience (supplier-on-time delivery, forecast accuracy), and profitability (working capital, margin impact). These metrics ensure digital investments translate into measurable business value.

Q. Looking ahead over the next five years, what emerging technologies and supply chain innovations will have the greatest impact on India’s engineering and manufacturing ecosystem? What strategic priorities should Indian manufacturers pursue today to build globally competitive, intelligent, and future-ready operations?

The next five years will be defining for Indian manufacturing. Policy momentum — from PLI incentives and the India AI Mission to Make in India 4.0 and the Digital Personal Data Protection Act —is converging with rapid advances in AI, IoT, digital twins, and robotics. This window of opportunity will not remain open indefinitely. PwC estimates AI could add $135–150 billion to India’s manufacturing MSMEs by 2035, but that value will accrue only to organizations building the right capabilities now.

Three technology clusters will shape the sector’s trajectory.

First, multi-agentic AI and autonomous supply chain operations, where specialized AI agents coordinate demand sensing, replenishment, logistics, and supplier collaboration at a speed traditional planning cannot match.

Second, digital twins that model plants, networks, and logistics flows in real time, enabling predictive maintenance and continuous optimization — already visible in early adopters like Mahindra.

Third, Generative AI embedded into enterprise workflows, transforming procurement analytics, supplier negotiations, regulatory documentation, and risk intelligence.

For Indian manufacturers, the priorities are clear: modernize cloud ERP, elevate data governance to a board-level agenda, build AI fluency within the workforce, and position India as a resilient, quality-certified alternative in global supply chains. Technology and policy support exist; what remains is disciplined execution and strategic clarity.

About The Interview

Santanu Boral is a seasoned Solution Architect with over 25 years of experience in supply chain and life sciences, specializing in enterprise architecture, digital transformation, and automation. A leader of the Forbes Technology Council and a published researcher with IJERT, he holds a Post-Graduate Certificate in Senior Management from IIM Kolkata and studied master seller prog at D’Amore-McKim School of Business at Northeastern University in the USA. At Cognizant USA, he works as an Associate Director, leading supply chain solutions for global life sciences
clients, driving complex digital transformations across the globe. He is recognized as a champion of multi-agentic AI and has published thought-leadership articles on AI-driven inventory optimization and the future of healthcare supply chains.