McKinsey & Company is urging pulp and paper packaging mills to use artificial intelligence as a practical tool for reducing costs and improving competitiveness in a slower-growth, more volatile market. The consultancy argues that traditional cost programmes are no longer enough for an industry facing weaker demand, volatile input prices, oversupply in some grades and increasing pressure from lower-cost regions.

The global pulp and paper sector has been affected by several shocks, including the pandemic, war in Ukraine, geopolitical instability, tariff pressure and changing fibre availability. At the same time, post-pandemic e-commerce growth has normalised, consumer demand has softened and customers in consumer packaged goods are reducing inventories and placing more volatile orders. These shifts have made mill planning, utilisation and cost control more difficult.

McKinsey notes that the industry’s historic model was built around high asset intensity, strong variable margins and maximising throughput to absorb fixed costs. That model is now under pressure. Fibre, energy, chemicals, logistics, labour and maintenance costs are all creating margin strain, while overcapacity in some segments is reducing pricing power.

For pulp and paper packaging mills, AI is becoming less about experimentation and more about finding the true cost-optimal operating point of the entire production system.

The consultancy recommends a model it calls “site sprints”: fast, cross-functional interventions that bring together operations, procurement, energy management and data science. Instead of running isolated cost-saving projects, mills analyse fibre use, machine performance, energy systems, indirect spending and production planning as one connected system.

This integrated approach is important because changes in one area can affect another. A fibre substitution, for example, may reduce raw material cost but require changes in energy use, chemical dosing or machine settings to preserve product quality. AI can help identify these trade-offs and guide decisions that lower total cost without weakening specifications.

McKinsey highlights several areas where AI can create value. Advanced recipe management can optimise fibre, energy and other variable inputs across grades. Predictive models can identify lower-cost material combinations that still meet performance requirements. Manufacturing analytics can improve yield, reduce web breaks, stabilise runnability and reduce waste. Energy optimisation can align steam, power and utility systems with production planning and market price signals.

• Fibre efficiency: AI can help reduce losses and improve yield from source to finished web.
• Energy optimisation: digital models can support smarter steering of steam, power and utility systems.
• Procurement intelligence: generative AI can support supplier analysis, category redesign and indirect spend control.

According to McKinsey’s example, one global pulp and paper company deployed site sprints across its plant network in under two years, achieving cost savings of 8% to 20% at individual mills and generating several hundred million dollars of impact. The company is now said to conduct multiple site sprints annually, suggesting that the model can become an ongoing performance discipline rather than a one-off transformation exercise.

For packaging producers, the implications are significant. Containerboard, cartonboard and other paper-based packaging grades are under pressure to deliver competitive pricing while supporting sustainability and supply reliability. Mills that can reduce fibre waste, energy consumption and downtime will be better positioned to compete against producers with structural cost advantages in Latin America and Southeast Asia.

Generative AI is also expected to help with information-intensive processes such as supplier analysis, demand planning and production optimisation. When combined with traditional AI and operational expertise, these tools can support faster, more consistent decisions at scale. This is particularly valuable in mill environments where large amounts of process data already exist but are not always used effectively.

The next phase could go even further. McKinsey suggests that robotics and physical AI may eventually unlock more advanced mill automation. For now, however, the immediate opportunity lies in using data and cross-functional teams to reduce cost baselines, improve resilience and create more stable performance in a difficult market.

For the packaging value chain, the message is clear: digital transformation is no longer optional. As demand patterns become more volatile and input costs remain unpredictable, AI-enabled cost optimisation could become a decisive advantage for pulp and paper packaging mills seeking to protect margins and strengthen long-term competitiveness. Source: user-supplied article. :contentReference[oaicite:0]{index=0}

Image concept: a modern pulp and paper packaging mill control room showing AI dashboards, fibre flow data, energy optimisation graphics, operators monitoring paper machines and reels of containerboard moving through production.