Managing the trade-off among packaging waste, throughput performance, and postal fees by optimizing the use of e-commerce packaging machines (OptiPack)

• Applicants: Professor Dr. Nils Boysen; Professor Dr. Dirk Briskorn
• Subject Area: Operations Management and Computer Science for Business Administration
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 569727205

A surge in packaging waste accompanies the rapid growth of online retail. The OptiPack project seeks to reduce the waste footprint by optimizing the setup and operations of automated packaging machines used in e-commerce. Fully automated packaging machines, capable of processing up to 1,000 products per hour, are supplied with goods via a conveyor belt and bands of cardboard packaging material via a feeding shaft. The cardboard is folded over the product, cut off in flow direction, sealed, labeled, and forwarded to the shipping area. Packaging machines and their usage are associated with a complex trade-off. Most packaging machines provide multiple cardboard bands of varying widths to reduce the packaging waste for differently sized products. The more alternative cardboard bands are available for a packaging machine, the better the fit for a given set of orders and the smaller the packaging waste. Analogously, more cardboard bands reduce postal fees for forwarding packed orders with a postal service provider. On the other side of the trade-off are the throughput performance of the machine and the investment costs. More cardboard bands fed into a machine increase the number of switches among them, and a more elaborate machine with additional feeding shafts increases investment costs. To derive suitable decision support, the OptiPack project defines and analyzes the following decision tasks and derives intelligent optimization algorithms for their solution: Crucial long-term decisions are related to selecting a suitable number of cardboard bands and specifying the width of each band. During operations, there is the option to alter the inflow of products on the packaging machine’s inbound conveyor, which is an important lever to reduce the throughput loss for switching among cardboard bands. Furthermore, a simple mechanical device on the inflow conveyor can manipulate the product orientation by flipping their direction, influencing waste, postal fees, and throughput loss. We aim to support these decision tasks (and their manifold interdependencies) for different types of packaging machines with intelligent decision support. This way, we enable packaging machine users to reduce their packaging waste footprint without neglecting their further aims.

• DFG Programme: Research Grants