Cloud-based 3D printing service allocation models for mass customization

Due to easy access, faster production, and on-demand printing, the applications of 3D printing are proliferating and reaching the mass customization level where individuals can print their self-design products. Many service providers emerge and supply 3D printers to allow mass customization within their specific service areas. Along with the technological advancement, these distributed 3D printers can be integrated and shared via a cloud-based platform for the manufacture of customized products in a dynamic and cost-effective environment. However, the 3D printing service allocation is a key challenge for the platform, especially when customers have specific time requirements for production. This paper investigates a 3D printing service allocation problem with optimization-based and real-time allocation strategies. The workflow of 3D printing service allocation is first analyzed in a cloud-based platform. Subsequently, a binary integer linear programming model is developed to allocate the 3D printing services to tasks given global demand information. A real-time allocation model is then developed on a first-come-first-serve basis to match 3D printing services with tasks. The objective is to maximize the net revenue of the platform. Numerical studies are conducted to compare the performance of two proposed models. The results show that the optimization model can increase the profit by approximately 100% on average, which implies the optimization-based strategy is superior to the real-time strategy in a metropolitan city; and because of the penalty mechanism, the acceptance rate can be increased by 5%. The results also show that the operators can find the optimal ratio of supply and demand to maximize the net revenue of the platform through historical data.