Selecting manufacturing partners in push and pull-type smart collaborative networks

The idea of Collaborative Manufacturing, also known as the Production Networks or Social Manufacturing, has been around for more than 25 years. It is a production concept based on non-hierarchical collaboration among enterprises often referred as Virtual Enterprise (VE). Despite many scientific research and projects with this topic, it is difficult to find an example of fully operational non-hierarchical production network anywhere in the world. However, that fact could be changed very soon. Namely, the new industrial revolution, called Industry 4.0, encourages industrial enterprises to adopt information-communication technology (ICT) and Internet of Things (IoT) into their production systems, thus creating Cyber-Physical Production System (CPPS). From the aspect of production networks, CPPS represents crucial infrastructure, or a missing link between enterprises. Now, with CPPS in place, non-hierarchical networking and collaboration becomes possible through Smart Collaborative Production Networks. In this research, the concept of information system for Smart Collaborative Production Networks was developed and called ‘VENTIS’. Although the idea of the concept is to manage the collaboration inside Virtual Enterprise, in this research, a special focus has been put on manufacturing planning phase in which optimization problem known as the Partner Selection Problem (PSP) occurs. Since the PSP in manufacturing phase is far more complex than partner selection during the collaborative product development phase, new research premises regarding the Virtual Enterprise type have been set. Two types of Virtual Enterprise business models – Push-type and Pull-type – have been defined in this research. If VE is Push-type, HUMANT algorithm is used to solve PSP that occurs in that case. If VE is Pull-type, a special procedure, inspired by phenomenological reduction, has been established in which set of a priori created VEs is compared with theoretically ‘the best’ VE and theoretically ‘the worst’ VE. Enterprises’ data of production network from Dalmatia (Split-Dalmatia County, Croatia) is used as a Case Study to present ‘VENTIS’ concept and to present the procedure for creation of sustainable Virtual