Skip to main navigation Skip to search Skip to main content

An MILP model for safe multi-floor process plant layout using the domino hazard index

  • Jude O. Ejeh
  • , Songsong Liu
  • , Lazaros G. Papageorgiou*
  • *Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

In this paper, an optimisation-based approach to obtain safe multi-floor process plant layout designs using the domino hazard index (a sub-index of the integrated inherent safety index) is presented. A mixed integer linear programming (MILP) model is proposed to obtain the economically optimal multi-floor layout design considering connection by pipes, horizontal and vertical pumping of process fluids, purchase of land, fixed and area-dependent construction of floors, the financial risk associated with hazardous events and their escalation potential, and the installation of passive protection devices. Hazardous events such as pool fires, jet fires, flash fires, fireballs and blast waves resulting from explosions are considered using a novel and more realistic estimation of safety distances between equipment items. A bi-objective optimisation problem is also considered, minimising the layout costs and the total domino hazard index values for the plant, adopting the ϵ-constraint method. The proposed model is then applied to an 11-unit case study susceptible to each of these hazardous events, obtaining results with the optimal layout and protection device configurations in a relatively short amount of time.

Original languageEnglish
Pages (from-to)137-165
Number of pages29
JournalProcess Safety and Environmental Protection
Volume148
DOIs
StatePublished - Apr 2021
Externally publishedYes

Keywords

  • Domino hazard index
  • Mixed integer linear programming (MILP)
  • Multi-floor process plant layout
  • Multi-objective optimisation
  • Safety

Fingerprint

Dive into the research topics of 'An MILP model for safe multi-floor process plant layout using the domino hazard index'. Together they form a unique fingerprint.

Cite this