OpenVendhacia un ecosistema abierto para el vending en la era de Internet de las Cosas

Abstract

This work describes how to apply Internet of the Things (IoT) paradigms to unattended point of sales such us vending machines. In order to face the 4th industrial revolution and contribute to mitigate climate change, this thesis proposes to accelerate the digital transformation of vending sector. To this end, each vending machine is modeled on the Internet as an online shop, allowing end-users to buy and pay from their mobile. Bear in mind this simple scenario, pervasive web and mobile technologies are analyzed to build up a low cost and easy-to-use solution. Taken into account that vending sector is very fragmented and operates at very low margins, this thesis also looks for operational efficiencies which may justify the investments needed to connect all vending machines to Internet. In this sense, a low-cost solution is proposed based on open hardware devices and a software as a service (SaaS) model. The solution has to be very easy to operate to minimize maintenance cost and must be ready for Big Data. To achieve these objectives, a private cloud architecture is designed to automatize the provisioning and management of devices. On the other hand, an alternative mobile payment system is proposed that seeks the simplicity and immediacy of payments with coins. The system must be universal, i.e., the end-user does not have to be affiliated to any financial institution, mobile operator or device manufacturer in order to pay with his mobile phone in any of the vending machines connected to the Internet. The design of the solution must ensure that the consumer is in front of the machine when the purchased product is dispensed. Under these premises, OpenVend has been implemented. OpenVend is a comprehensive open source & hardware solution that improves the operational efficiency of vending operators and enables e-commerce transactions from consumers' mobile phones by modeling the machine as an online shop. The outcomes of this research can be extended and applied to different sectors and scenarios where it is necessary to pay for goods and services in an unattended manner, or simply to eliminate queues. For example: paying a parking lot, renting a bicycle, paying in restaurants and supermarkets, etc. In any of these scenarios, the digital modeling of a complex system or service (digital twin) can be instantiated in the cloud automatically through the plug-and-play mechanism presented on this work. Once the digital twin is instantiated, the end-user can interact with the physical system through his digital twin. OpenVend has been built with security by design and therefore guarantees the privacy of end-user data.