Submitted Abstract
The rise of new information technologies offers the waste recycling industry new prospects for reducing pollution, increasing the quality of existing services, and creating new performance indicators for stakeholders in the sector. A perfect example is the SWC market, which for several years has been looking for mechanisms to automatically adapt collection routes to changing needs and constraints, rather than relying on fixed schedules. This project is positioned in this market and focuses on collecting waste from professional customers, whose properties are different from those of homogeneous private individuals. In addition to well-known route optimisation models deployed in business applications, the exploitation of sensor network technologies to smart cities is becoming increasingly relevant and is expected to increase by 248% by 2026. However, solutions that really benefit from this modern technology are rare, and existing ones usually do not consider end-user and business priorities in their decision (i.e. constraints and needs). SWAM aims to propose and validate a novel SWC platform relying on multi-objective optimisation processes, in order to combine business, customer, and operational criteria – including data generated by new fill-level sensor technologies integrated into waste bins. This will be based on a novel multi-channel analysis approach, combining opportunistic data flows with customer and business priorities. In this context, LIST has partnered with PoL to go further in the digitalisation process around existing professional SWC solutions. The SWAM decision-making platform will comprise two fundamental systems. Firstly, a Data Management System (SWAM-DMS) will collect and process multiple data flows provided by PoL and external services (e.g. road traffic), including: (a) waste bin filling levels, regularly transmitted by new ultrasonic sensors to be deployed as part of the project; (b) business data flows composed on the one hand of data imported from PoL’s ERP (e.g. customer inventory) and on the other hand of needs and constraints reported by PoL’s customers and defined according to their socio-professional categories. Secondly, a Business Process Optimisation System (SWAM-BPOS) will consider outcomes generated by SWAM-DMS as input to generate multi-objective models for dynamically optimising collection routes and targeted indicators distributed through a control centre (fleet management, clients’ profiles) and a route navigation assistant. The effectiveness of the solutions to be developed within SWAM and the research objectives will be validated through real-life deployment. The assessment of the business implementation potential will be carried out throughout the project in order to develop future opportunities both for LIST, PoL, and potentially for other companies in the “collection of non-hazardous waste” sector (NACE 38.11).