Celebrating the 10th Volume of the Journal of Simulation
As we enter 2016 ,the Journal of Simulation (JOS) embarks on its tenth volume. We both remember a discussion with Professor Ray Paul in 2004 in which he suggested that we might consider creating a new journal in our area with the Operational Research Society (ORS). After some thought (and anxiety!) we proposed the concept of JOS to the ORS and Palgrave Macmillan. The time has flown since the inaugural issue was published in December 2006. That issue was announced with a celebration at the Winter Simulation Conference in Monterey. Throughout the preceding nine volumes the journal has developed its presence, attracting papers from around the world and achieving a Thomson-Reuters ISI rating in the minimum time possible. Among the many authors who have contributed to the pages of JOS have been world leaders in their respective fields of interest. The journal continues to publish a broad range of papers focussing on the latest developments and applications in simulation.
To mark the tenth volume, the current editor-in-chief (Christine Currie), invited the founding editors (ourselves) to bring together this special issue based on the papers published in the first nine volumes of JOS. We set about the task by identifying one key paper from each volume. This was no easy task. The resulting set of papers demonstrate the breadth of topics that the journal has covered. As such, the papers that follow provide an interesting account of the history, theory and practice of simulation.
Beyond our inaugural and extended editorial, the first volume opened with a paper from Pidd and Carvahlo (2006) entitled ‘Simulation software: not the same yesterday, today and forever.’ The paper is reproduced in this celebration issue. Pidd and Carvahlo map out the parallel developments in computing and discrete-event simulation and they postulate that modern simulation packages have not kept pace with developments in computer software. A particular area of deficit was in component-based development. They address this issue by describing a prototype component-based discrete-event simulation package. We would no doubt argue that the gap between developments in computing and simulation software has persisted through to today in nature if not in form.
Volume 1 concluded with a special issue on conceptual modelling (issue 3). This issue was at the start of a new wave of research into this important topic. In the issue, Arthur and Nance (2007) identify the close links between software requirements engineering and conceptual modelling for simulation. The issue also contains the first winner of the Tocher Medal (see below), Kotiadis’ (2007) paper ‘Using soft systems methodology to determine the simulation study objectives.’
Issue 3 of Volume 2 in Hollocks (2008) provided the opportunity to celebrate 50 years since the first simulation software, produced by KD Tocher at the United Steel Company in the United Kingdom. We reproduced two of Tocher’s early papers on simulation and reviewed his groundbreaking book from 1963, ‘The Art of Simulation.’ The main feature of that issue was Hollocks’ (2008) paper describing in detail the life and work of Tocher. Hollocks had spent many years working on simulation with Tocher in the steel industry and so was in a unique position to give an account of Tocher’s contribution to our field. The paper is based on Hollocks’ (2008) keynote presentation at the 2008 Operational Research Society Simulation Workshop. A highlight of that event was welcoming Tocher’s widow, Charlotte, and his two children to the keynote presentation at which we also announced the founding of the Tocher Medal, awarded biennially for the best paper in JOS.
For volume 3 we have identified quite a different paper. Boer et al (2009) carried out a survey on distributed simulation in industry. Their paper reports on the findings of their survey. What they found is that industry is underdeveloped in its use of distributed simulation. They identify a range of reasons as to why this might be the case. As with Pidd and Carvahlo’s paper, Boer et al again highlight the lag between developments in computing and, on this occasion, progress with simulation in industry.
As we reached 2010 and Volume 4 we published the first of our special issues on agent-based simulation, edited by Raymond Hill; the second was published in 2013. We did this in response to a growing interest in this field and, indeed, since then JOS has attracted many papers on the topic of agent-based modelling. The special issue opened with Macal and North’s (2010) tutorial on agent-based modelling and simulation. This paper, which is reproduced here, is an excellent introduction to the topic and remains the most downloaded paper from the journal’s archive. The issue also contained an interesting discussion on the role of discrete-event simulation versus agent-based simulation. This paper, by Siebers et al (2010) is a summary of a lively panel discussion on the topic at the 2010 Operational Research Society, Simulation Workshop.
From Volume 5 we have chosen Beck’s paper on modelling passenger flows in Heathrow Terminal 5. Beck (2011), a member of the operational research team for British Airways, describes how they used simulation to help design the new terminal. JOS has always had an interest in publishing papers of this nature that demonstrate the practice and application of simulation, and this is a particularly significant example as it addressed a key component of the UK’s national infrastructure that was designed to handle in excess of 30 million passengers a year.
Volumes 1–5 contain a set of papers that reflect on how simulation projects can be performed better. Although these papers are not reproduced in this commemorative issue, we think they are worth mentioning. In Volume 1, Chick (2006) describes his experience of simulation projects in his paper ‘Six ways to improve a simulation analysis.’ Meanwhile, Salt (2008) adopts the opposite approach in Volume 2, looking at what can go wrong in a simulation study in ‘The seven habits of highly defective simulation projects’. A further paper appears in Volume 5, when Banks and Chwif (2010) give us ‘Warnings about simulation’.
Volume 6 key papers also include significant practice papers on container terminal efficiency and pharmaceutical supply chains, as well methodological papers on (outer) space operations and hypermodelling. However, we felt that the contribution made by Yavari and Roeder (2012) on metrics for model enrichment was the most significant paper of the year. The work introduces measures to capture ‘effective’ simulation models. They introduce the notion of ‘Enrichment Level’ that allowed alternative models produced in a project to be compared systematically. Given the wide variety of models that could be produced in a project, their metric gives a useful foundation to justify the choice of model to be simulated on a quantitative, rather than arbitrary, basis.
Sargent’s (2012) extended verification and validation of simulation models paper was one of two key papers we considered in Volume 7. This is a major paper that comprehensively addresses a wide range of issues in verification and validation, as well as presenting a recommended procedure for model validation. However, on balance, we selected Tolk et al’s paper on reference modelling for simulation. There has been discussion for many years on the closeness of systems engineering to simulation Tolk et al. (2013). Their paper brings the two areas together to present a framework that introduces systems engineering concepts across all aspects of modelling and simulation and, perhaps, reflects the benefits of closer links between OR/MS and software engineering.
The results of a lively panel discussion on discrete-event simulation and other simulation paradigms, product and process patterns for agent-based modelling and simulation (Brailsford, 2014), and Kleijnen’s (2014) comprehensive survey of simulation optimization via Kriging and bootstapping were all important papers in Volume 8 for different reasons. Zaman et al’s (2013) paper really stood out as a major example of simulation practice. Their work deals with maritime safety in the Strait of Malacca, the longest strait in the world and a high-risk area for navigation. Carrying around 40% of world trade, it is critical to assess the risk of collisions. They propose hazard identification and risk evaluation steps as part of formal safety assessment methods for ship collisions. This shows how modelling and simulation can have an impact (by avoiding physical impact!) at a potentially global level.
In our most recent Volume three papers were selected. Multi-echelon production inventory optimization is extremely important to large volume factories, and Güller et al (2015) clearly show how simulation can be used effectively to support that optimization. Lin et al’s (2015) excellent paper discusses the simulation model risk caused by estimating input distributions from real-world data and proposes the first single-run method for quantifying this input uncertainty. Finally, we selected for Volume 9 Smith and Harper’s (2015) paper that reported on the sustainability of a rural community health centre in North India. This is remarkable paper as it shows rigorously how simulation can be used to aid the planning of vital health services in remote, under-developed regions of the world.
It has been extremely difficult to select (and even shortlist) the key papers from the first 10 years of JOS. These papers represent major technological advances, methodological advances, new paradigms, industrial impact and significant examples of real simulation practice across the world, as well as reflections on the roots of our field. A journal is only as good as its papers and its peer-review process.
The papers we have published over the last decade have been of very high quality and the authors would like to thank the other authors and the reviewers for all their hard work in creating such excellent work. The authors would also like to thank the others who have joined us on the way, especially John Fowler, Loo Hay Lee, Christine Currie and the Associate Editors, and for the Operational Research Society and Palgrave Macmillan for their ongoing support. Here’s to the next 10 years!