Innovative Applications of O.R. Macroscopic attraction-based simulation of pedestrian mobility: A dynamic individual route-choice approach David Canca a,⇑ , Alejandro Zarzo b , Encarnación Algaba c , Eva Barrena c a Department of Industrial Engineering and Management Science, School of Engineering, University of Seville, Av. De los Descubrimientos s/n, 41092 Seville, Spain b Department of Applied Mathematics, School of Industrial Engineering, Technical University of Madrid, c/José Gutiérrez Abascal, No. 2, 28006 Madrid, Spain c Department of Applied Mathematics II, School of Engineering, University of Seville, Av. De los Descubrimientos s/n, 41092 Seville, Spain article info Article history: Received 18 May 2012 Accepted 25 May 2013 Available online 4 June 2013 Keywords: Simulation Pedestrian mobility Dynamic route choice Route assignment Transportation planning abstract This paper presents a dynamic distribution and assignment simulation model based on discrete time sim- ulation techniques and dynamic route assignment for planning, engineering design, and operation anal- ysis of big exhibition events from a pedestrian circulation perspective. Both, the distribution and assignment stages are incorporated in an interlaced way with a dynamic behavior along a specific time horizon. In the proposed model, the individual route choice is dynamically determined as consequence of facilities attractiveness and network congestion. Therefore, in contrast with other simulation approaches, it does not require the usual origin–destination trip matrices to describe the transportation demand or the specification of different paths to be followed by visitors. This modeling approach turns out to be very appropriate for the simulation of these big exhibition events where each visitor usually has multiple and a priori unordered destination choices after entering the scenario. The main aim of the proposed model is to obtain measures of pavilions occupation, queue sizes and links density at the various facilities rather than the exhaustive tracking of individual movements or the precise description of the interaction between different entities. So, the use of pure microscopic con- tinuous models is not needed. This approach can be viewed as a macroscopic simulation based on indi- vidual route-choice behavior of entities (microscopic level). As illustration, here we show how the model works when dealing with the pedestrian mobility in the International Exhibition of Zaragoza held in 2008, where it was used as a planning and operation analysis tool in the preliminary stages. Ó 2013 Elsevier B.V. All rights reserved. 1. Introduction One of the most important objectives of the mobility and behavioral pedestrian studies is to evaluate the effect of a proposed policy on the pedestrian facilities before its implementation. Obvi- ously, to carry out and develop these policies without a previous and reliable analysis might lead to a very costly trial and error practical implementation. This is especially true when dealing with international exhibitions and fairs because they are characterized (Zhou et al., 2009a) by a big surface with multiple types of pedes- trian facilities (pavilions, restaurants, transportation facilities, landscape zones, etc.) and by a random behavior of visitors, with different and consecutive destinations after entering the exhibition scenario. For this kind of analysis, simulation has been a useful tool, both in a standalone implementation and in conjunction with optimization techniques. 1.1. Background There have been many simulation studies applied to the analy- sis of pedestrian movements in different scenarios and locations. During the 1960s and 1970s a handful of pedestrian demand mod- els were developed with the aim of forecasting pedestrian flows in central business and commercial district areas. These models were developed with a structure similar to the so-called four-stage transportation planning models (Schofer and Levin, 1967). They incorporate three fundamental aspects: trip generation (based on land-use characteristics), trip distribution (Waddell et al., 2007; Zhou et al., 2009b) and assignment over the corresponding net- work following an a priori user equilibrium model approach (such as e.g., the Wardrop’s first and second equilibrium principles, War- drop, 1952). Besides these aspects, all these studies have an impor- tant feature in common: to achieve their goal, they do not require any microscopic description of the pedestrian two-dimensional movements and interactions. Assignment plays a central role in the quantitative analysis of transportation systems. Since the work of Beckmann et al. 0377-2217/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ejor.2013.05.039 ⇑ Corresponding author. Tel.: +34 954486197; fax: +34 954487329. E-mail addresses: dco@us.es (D. Canca), alejandro.zarzo@upm.es (A. Zarzo), ealgaba@us.es (E. Algaba), ebarrena@us.es (E. Barrena). European Journal of Operational Research 231 (2013) 428–442 Contents lists available at SciVerse ScienceDirect European Journal of Operational Research journal homepage: www.elsevier.com/locate/ejor