Maritime transportation planning under carbon intensity constraints for sustainable supply chains

PRESSER, DEMIAN J.; CAFARO, DIEGO C.; GROSSMANN, IGNACIO E.; MISRA, PRATIK; MEHTA, SANJAY

Buenos Aires

Congreso; WCCE11 - 11th WORLD CONGRESS OF CHEMICAL ENGINEERING; 2023

Asociación Argentina de Ingenieros Químicos

Assessing sustainability in supply chains is becoming imperative to massively deliver consumer goods, resources and energy to environmentally concerned markets [1]. Carbon intensity (CI) stands for an indicator weighting the amount of carbon emissions per unit of output produced by logistics and processing operations, and represent an essential measure of environmental performance. Achieving world climate goals while remaining competitive requires industries to focus their efforts on managing carbon emissions of their operations and keep track of the carbon footprint of products along their value chain. Improving this indicator requires operations to be carried out in alternative, less economical means. CI control strategies become even more challenging when applied to chemical and/or bulk products operations because mixing streams coming from different sources may have a relevant impact on meeting specifications. Maritime transportation is a common task in most global supply chains, with a strong impact on carbon emissions. The selection of fuels for the ships and their transportation speeds can be decisive in ensuring product specifications on CI at the delivery points. Furthermore, meeting demand overseas may require the use of different ships, tanks and sourcing strategies for every receiving port, usually requiring revaluation of product properties after blending in any vessel. The maritime inventory routing problem under CI constraints aims to optimally plan ships transportation while also accounting for carbon emissions, and can be stated as follows. Given: (a) a set of source nodes producing chemicals or bulk materials at known yield rates and associate emissions; (b) a set of receiving ports with known demand rates and different CI requirements; (c) a fleet of ships of different capacities, fuels and emission rates according to their travel speed; and (d) a set of tanks placed at every port in the network with different capacities; we seek to determine: (1) the sequence of ports to visit (routes); (2) the loading and unloading amounts from every ship at every port; and (3) the duration of ship trips and their corresponding carbon emissions, in order to minimize the total cost. In this work we propose a novel mixed integer nonlinear mathematical programming model (MINLP) based on a hybrid discrete-continuous time framework, seeking for the most economical plan of material supplies of on-spec products to customers. In contrast to previous contributions [2], we include nonlinear functions for the accurate assessment of emissions in terms of the ships speed. Results show that managing CI indicators can significantly impact overall service levels and modify optimal plans purely based on economic criteria. Finally, we present an efficient decomposition strategy that allows solving large instances of the problem in the long-term, based on a receding horizon framework. References1.Rajeev A., Pati R.K., Padhi S.S., Govindan K. (2017). Evolution of sustainability in supply chain management: A literature review. Journal of Cleaner Production, 162, 299-314. 2.Presser D., Cafaro D.C., Grossmann I.E, Misra P., Mehta S. (2022). Mathematical programming models for the optimal management of carbon intensity indicators in global supply chains. Submitted to Computers & Chemical Engineering.