A Hybrid Slot-Based/General Precedence Approach for Planning Crude Oil Supplies

PEDRO C. PAUTASSO; VANINA G. CAFARO; JAIME CERDÁ; DIEGO C. CAFARO

Computer Aided Chemical Engineering

Elsevier

Año: 2017; p. 1321 - 1326

This work introduces a mixed-integer nonlinear programming (MINLP) formulationbased on a hybrid approach, combining the potentials of slot-based and generalprecedence continuous-time representations. We use general precedence sequencingvariables to coordinate incoming/outgoing flows to/from every tank, and simultaneouslyaddress the scheduling of a long-distance pipeline supplying crude oil batches fromharbour to refinery tanks, following a slot-based scheme. The model is able to preciselymonitor key component concentrations keeping oil properties within admissible ranges.One of the most important decisions is how to manage oil shipments so as to improvelogistics efficiency. Results show that the use of slot-based frameworks combined withgeneral precedence variables for sequencing tasks in linking resources yields orders-ofmagnitude savings in the computational effort to reduce costs and find optimalsolutions.