Journal of Simulation (2009) 3, 1–2
r 2009 Operational Research Society Ltd. All rights reserved. 1747-7778/09 www.palgrave-journals.com/jos/
Editorial Simulation software: evolution or revolution? Journal of Simulation (2009) 3, 1–2. doi:10.1057/jos.2008.25
At the landmark 50th Operational Research Society Conference (York, 9–11 September 2008), we had the opportunity to run a discussion session on simulation software. Rather ambitiously called ‘Simulation Software: Evolution or Revolution’, the idea was to continue the debate started in our forward-looking survey that appeared in the first issue of JOS (Taylor and Robinson, 2006) by focusing on simulation software. After a brief presentation the 25 or so attendees were invited to reflect on four areas of simulation software: simulation technology, simulation applications, simulation practice and simulation experimentation and analysis. The attendees were both researchers and practitioners from an OR/MS background. The results of these are presented in Tables 1–4 without bias—we leave it to the reader to decide
upon the relative merits of each for themselves. We encourage others to hold similar sessions (as an occasional alternative to a panel) and to submit the results to JOS. The process and the outcomes can be quite surprising! Our third volume marks a shift from three issues to four. In this issue Boer, de Bruin and Verbraeck report on a survey of distributed simulation in industry. Van der Zee and Slomp consider a case study on the alternative use of simulation for training industrial workers in new working procedures. Finley et al present an efficient computationbased algorithm for effecting round-robin service in discrete-event simulation. Onggo discusses a multi-faceted representation of conceptual models to initiate discussion
Table 2 Simulation applications K
K
K K K
Table 1 Simulation technology issues K K
K K K K K
K K
K K K K K
K K K K K
K K K K
Better facilitation of model/sub-model reuse Better links between COTS simulation packages and physical simulators Better links to process maps Better support for group use of models Better support for optimization Better use of graphics and animated sequences Books describing how COTS simulation packages can be used in different domains Cheaper packages Conversion between ‘drag and drop’ models and Java/ C þ þ program code Discrete-event simulation and system dynamics conversion Easy to use agent-based simulation Easy web-based deployment for non-expert Easy interoperability Embedded decision support algorithms (run dependent on decision points) Intuitive use More niche modelling/embedded models Off-the-shelf support for object-oriented model creation Portability of models between COTS simulation packages Provision of run time and development licences for COTS simulation packages Standard for comparing simulation techniques Standardized reuse of object-oriented simulations Support for domain templates User friendly agent-based simulation
K K K K
K K
K
Amend CAD drawings and import into COTS simulation packages Applications to third world through non-government organizations and charities Appropriate graphic libraries (cross package) ‘Beginners’ documentation Common terms and jargon Distributed applications Financial applications Integrated lifecycle simulation Manufacturing applications: material flow, constraint analysis Modelling human factors issues Open platform availability to run widest platform availability User friendly patient flow simulations
Table 3 Simulation practice K K K K K K K K K
K
K
K
Better links with lean processes Better support for data collection Conceptual modelling for agent-based simulation Conceptual modelling frameworks Configuration of COTS tools Deployment into user groups (eg engineers) Develop use of probabilistic sensitivity analysis Improved validation techniques Integration of conceptual modelling and pre-modelling tools Integration of simulation modelling methods into enterprise development methods Research into comparative cost of model development by COTS simulation package or by object-oriented programming Simulation as facilitation/problem structuring
2 Journal of Simulation Vol. 3, No. 1
Table 4 Simulation experimentation and analysis K K K
K K K
K K
K K K K K K
Better integration of optimization with simulation tools Better visualization for experimentation Design and analysis of experiments training for practitioners (not just another statistics course) Guidance on scenario selection More options for exploring solution space ‘On-line’ selection/implementation of variance reduction techniques for simulation optimization Quick model development and easy to understand analysis Real-time experimentation to see likely effect of making different decisions Recommendations for number of replications Results management and analysis support Scenario management Sharing of simulation analysis via the Web User-friendly interfaces Wider experimentation support
between stakeholders and demonstrates its use with a case study in healthcare. Longo and Mirabelli attempt to develop an effective design of an assembly line for heaters production by taking into account work measurement, line balancing and ergonomic problems. Finally, Corriea and Viagas describe a conceptual model of a simulation for studying alternative measures to reduce the number of automobiles travelling every day to city centres. SJE Taylor and S Robinson
Reference Taylor SJE and Robinson S (2006). So where to next? A survey of the future for discrete-event simulation. Journal of Simulation 1(1): 1–6.