PhD Thesis

Towards the development of a decision-support tool for the production planning and campaign scheduling of biopharmaceutical facilities

Miguel Jorge Vieira2017

Key information

Authors:

Miguel Jorge Vieira (Miguel Jorge Vieira)

Supervisors:

Tânia Rute Xavier de Matos Pinto Varela (Tânia Rute Xavier de Matos Pinto Varela); Ana Paula Ferreira Dias Barbosa Póvoa (Ana Paula Ferreira Dias Barbosa Póvoa)

Published in

October 20, 2017

Abstract

One of the main challenges in the management of most industrial companies relies in the optimisation of its production systems while assuring the dynamic operational complexity and its key process economics. The case of the pharmaceutical industry has been the research context for several studies on production optimisation, aiming the effectiveness of its long drug development cycles and compound manufacturing activities. Moreover, the rising of the biotech sector in the production of highly effective drugs has placed the biopharmaceutical industry as a new scope for planning/scheduling decisionsupport methodologies. Therefore, the development of an effective modelling formulation to address the specific operational and financial constraints of these bioprocesses is the core of the proposed research, assessing the requirements for the optimal support to industrial decision-making. As detailed in this thesis, the research starts to discuss the overall model-based approaches to address optimal industrial planning/scheduling solutions focused on a review of related literature. Several challenges are identified centred in the model formulation, computational efficiency and large-scale implementation. A generic model formulation is then defined to address the management problem constraints of biopharmaceutical production planning. These complex biochemical processes target for flexible and cost-effective operations while dealing with long campaign processing times, intermediate product stability/quality issues and the high cost/low yield of manufacturing steps. A mixed integer liner programming model is developed and the model performance is studied through a set of industrially relevant cases/scenarios. Lastly, the framework for the implementation of an enterprise decision-support application is defined with a proof-of-concept study, assessing the decision-maker trade-offs (model detail vs computational time vs quality of solutions) and the industrial requirements. The work of this thesis highlights the benefits of mathematical optimisation in production planning/scheduling problems, following a systematic approach to real biopharmaceutical industrial cases for the implementation of a decision-support solution provided by an innovative model-based application.

Publication details

Authors in the community:

RENATES TID

101595310

Degree Name

Doutoramento em Líderes para Indústrias Tecnológicas

Fields of Science and Technology (FOS)

economics-and-business - Economics and business

Publication language (ISO code)

eng - English

Rights type:

Only metadata available

Institution name

Instituto Superior Técnico

Financing entity

Fundação para a Ciência e a Tecnologia