Long regarded as a cost center, the supply chain has become a strategic asset and a source of value. At Tech For Industry Show 2026 (Paris Expo Porte de Versailles), three manufacturers — Christophe (Groupe Nestlé), Ludovic (Renault Group) and Perceval (aerospace) — compared their roadmaps in the face of now-structural instability, around one horizon: the autonomous supply chain.

For Nestlé, the supply chain is no longer a "given" cost center but a source of added value that contributes to growth and consumer response. In aerospace, the shift is even clearer: "we have no sales problem," fleets are being renewed and all OEMs (Airbus, Boeing) are growing. The challenge becomes producing at the right cost and the right speed — and mastering the supply chain becomes the key link in the entire industry's performance.

Covid revealed structural fragility: many small players, little investment in tools, high skills turnover. A striking example: in 2025, a Pennsylvania factory, single-sourced on key components, burns down. The result: supply disruption, then panic buying — everyone over-stocks to protect themselves, creating a "demand bubble" that worsens the initial crisis. "We're truly subject to the butterfly effect; we see it every day."

Fragmented flows, customs barriers, geopolitical events: shockwaves multiply. A concrete aerospace example: at the start of the war in Ukraine, the ban on Russian titanium — the titanium of aerospace — forced the entire sector to urgently reconfigure toward other sources, with "first come, first served" panic movements.

Nestlé points to three other factors. Climate risk, increasingly dominant (coffee growing zones, agricultural yields). Cybersecurity: pooling storage zones exposes companies to external attacks — "we ended up with 70,000 pallets blocked," and even three KitKat trucks stolen through impersonation. And changing consumers: e-commerce now accounts for 30% of food volumes, with highly variable delivery cycles — from Carrefour every two days to 15-minute delivery in India. All within a staggering ecosystem: 337 factories, 1 billion products sold per day, 2,000 brands, 100,000 suppliers.

Renault expanded its scope to all tier-n suppliers: over 40,000 suppliers across every continent, ~4,000 tier-1, and over 22,000 distribution points. War in Russia (lost factories), daily cyberattacks through logistics "pipes," floods: this environment triggers an explosion of decisions to make at every moment (substituting a supplier, changing transport, rerouting ships). Hence the entry into the realm of decision intelligence, the engine of the entire digital transformation.

Before, Renault recalls, data was "unreliable, not available in real time, incomplete" — hence poor decisions, or no decision at all. The group therefore went data-driven across all of Renault Group: building an efficient supply chain also requires engineering and procurement data (optimizing transport by tariff grids, loading by part weight…). Silos between major functions were broken, data made available to as many as possible.

To extend visibility to tier-n, Renault deployed control towers — collaborative platforms where players share data to decide. On the IT side: a hyperscaler, a data lake, and a strong stance — storing as much data as possible (not just what's useful for an immediate use case), to later have the historical depth needed for predictive AI models (e.g. estimating the cost of an urgent transport recovery).

Nestlé stresses a complementary point: data usability depends on its interoperability. Without precise standards, no end-to-end vision. The approach: handling integrated chain segments (source-to-pay, order-to-cash, plan-to-produce) before orchestrating the full value chain. "A result is not just the sum of functional excellences; it's the fruit of orchestrating the whole value chain."

For aerospace, AI is a breakthrough: it frees people from transactional activities to focus on value creation. An envisaged example: an autonomous procurement agent handling the administrative part (sharing a forecast, confirming receipt, following up, negotiating dates) to leave humans the real value-added discussion with the supplier.

All share the "grail" of the autonomous supply chain, where each player decides autonomously thanks to a global view. Renault cautions: agentic AI makes this dream "within reach," but under two conditions. First, the data foundations: without a prior data-driven shift, agentic AI won't work — "there's an illusion with LLMs that we'll reconcile everything even if data isn't structured: no, that doesn't work." You need ontology and semantic layers (which AI can help produce faster). Second, don't give in to complexity: AI must push toward simplification, not complication, and guarantee reliable results without hallucinations.

Nestlé started early, in 2019, with predictive models for commodity prices (coffee, cocoa, sugar) for its hedging. After a year of benchmarking, the verdict: with good data foundations, the machine generates predictions 50% better than the human expert — prompting a redesign of both the model and the organization (experts move to improving the algorithms). Today, 1,000 agents optimize functional points and, increasingly, decisions across the whole value chain.

Nestlé's answer is nuanced: AI handles complexity, volumes and information speed; the decision remains human, except for very tactical activities already automated (e.g. choosing the cheapest, lowest-CO₂ air freight). Humans are, however, "excluded" from data collection and analysis — what took 3 weeks now takes 30 seconds (Nestlé's global procurement spend has been available in 10 seconds since 2019). A persistent barrier: human knowledge isn't fully integrated into databases, and hallucinations remain.

Sovereignty clashes with competitiveness: creating a dual source is costly. You must anticipate and model scenarios (a Taiwan conflict, for example), an area where AI helps push analysis far to ensure resilience at every tier.

At Renault, logistics accounts for 1–2% of a vehicle's carbon footprint (most coming from usage) — hence the priority on electrification, but also work on network design, load optimization and local ecosystems (around the Douai plant producing the electric R5 and Scénic, supplied by electric trucks: over one million kilometers, Everest Trophy). At Nestlé, sustainability is "intrinsic to strategy": working with farmers (cocoa, dairy in India and the methane issue), design to value integrating ESG impact from design, "zero water" factories, satellite monitoring of palm-oil deforestation. With a humbling lesson: after cutting 236,000 tonnes of plastic, recyclers stopped recycling certain products that became unprofitable — "sometimes you create a new problem trying to solve one." Hence the choice to work with NGOs rather than defend a position.

The disappearance of transactional tasks requires moving up the value chain: mastering data, dashboards, data analytics. New roles emerge (data steward, data analyst) and others become pivotal, such as the SPM (Supplier Performance Manager) in aerospace, who steers a supplier's industrial maturity with a 360° view (quality, commercial, production). A challenge made harder by post-Covid turnover.

The three manufacturers converge: the compass is value creation, not digitalization for its own sake. Aerospace is investing in a data-analyst department and leverages the Aero Excellence initiative (a common industrial-maturity framework for the whole sector). Nestlé combines process review with organizational review (designed in a "pre-digital" era), and identifies the real challenge: not resources, but the capacity to change and resistance to change, when starting from hundreds of thousands of employees. Renault stresses responsiveness: the Douai line runs Mégane, Scénic and Alpine together; when electric sales double, you must go from 600 to over 1,000 cars/day in weeks — "otherwise the customer will buy an available car, Chinese or otherwise."

What is an autonomous supply chain? A supply chain where each player decides autonomously thanks to a global view and real-time data access, supported by AI agents. It requires solid data foundations (structuring, ontology) before any agentic layer.

Why has the supply chain become strategic? Because it has shifted from a cost center to a source of value and competitiveness, and because structural instability (geopolitics, cyber, climate, consumer expectations) makes it the key link in industrial performance.

What is a logistics control tower? A collaborative platform that shares data among supply chain players (down to tier-n suppliers) to give each the visibility needed for their decisions.

Will AI replace human decision-making in the supply chain? Not at this stage: according to the panelists, AI handles complexity, volumes and data analysis, but strategic decisions remain human, except for very tactical tasks already automated.

This panel was recorded at Tech For Industry Show 2026, the B2B trade show for industrial transformation (Smart Manufacturing, AI, Supply Chain) at Paris Expo Porte de Versailles. Watch the full discussion above, and join us on May 19–20, 2027, Hall 7, for the next edition.