The increasingly rapid expansion of artificial intelligence in logistics also extends to the systems serving online commerce. To understand its current and future implications, the School of Management at the Politecnico di Milano, supported by Amazon, investigated the tangible effects of AI-based technologies in demand forecasting and inventory management processes. Through the analysis of real-world cases and interviews with companies operating within the e-commerce supply chain, the study highlights that the adoption of artificial intelligence significantly boosts operational efficiency, enhances customer service quality, reduces waste and, at the same time, supports the development of human skills.
The research shows that companies implementing predictive solutions and intelligent algorithms have achieved improvements in forecast accuracy, with gains ranging from 5 to 10 percent in already digitally mature organisations, and reaching up to 95 percent in less structured ones. This translates into more precise inventory management, a sharp decline in stock-out events and greater product availability in catalogues. On the operational side, companies also reported improved space management: around 40 percent stated they had recovered up to one fifth of their warehouse surface area, and in some cases, AI adoption allowed them to maintain the same operational capacity while halving the space used.
Transport operations have also benefited from predictive models. The average vehicle fill rate rose from 85 to 95 percent, resulting in a 30 percent reduction in the total number of trips. This has been accompanied by a marked decrease in waste: half of the surveyed companies reported a 50 percent drop in logistical inefficiencies, and approximately 70 percent eliminated most urgent deliveries. Meanwhile, customer experience has improved: 85 percent of businesses involved in the study observed a positive impact on service levels, 60 percent were able to increase stock availability and 40 percent expanded their product offering.
The effects of artificial intelligence are not confined to technical aspects. One of the most significant findings of the study concerns the human dimension. The introduction of AI did not lead to job cuts but instead encouraged companies to rethink roles and enhance new skill sets. Eighty percent of the companies reassigned staff to higher-value activities, and 40 percent have already launched digital upskilling programmes. As a result, new professional profiles are emerging, including data analysts, predictive planners and technicians specialised in the maintenance of robotic systems. Moreover, the integration of data across the entire logistics chain has enabled almost all companies to gain greater visibility over their flows and foster more effective collaboration across business functions.
The study by the Politecnico also provides a series of operational recommendations. Companies are advised to introduce artificial intelligence through pilot projects focused on demand forecasting and inventory control, to invest in structured skills development programmes and to incorporate sustainability metrics into their monitoring systems. For institutions, the proposals include offering tax incentives for the adoption of green AI-based technologies, strengthening technical and university training programmes related to data analysis and smart logistics, and promoting public-private partnerships. For workers, the message is clear: the future will demand increasingly advanced digital skills, making it essential to invest in professional development, including through training schemes co-funded by employers.
Within this context, the study presents the example of Amazon, which has accompanied its technological transformation with a focus on human capital. Since 2010, the company has invested over 20 billion euros in Italy, including four billion in 2023 alone, creating more than 19,000 permanent jobs. The Career Choice programme, which covers up to 95 percent of training course costs (up to a maximum of 8,000 euros), has been increasingly used by employees to access new career opportunities.
In 2025 alone, Amazon invested a further 700 million euros in automation and artificial intelligence for last-mile sorting centres, bringing the total investment to 1.4 billion over two years. Since 2019, more than 550 new technologies have been introduced in European logistics centres, enhancing over 60 different job roles. In Italy alone, 20,000 workers in the last-mile sector have been trained, and the number of technicians specialised in robotic maintenance has increased by 50 percent in just two years.
