Software selection for simulation in manufacturing: a review
Introduction
Despite the fact that the selection of suitable simulation software is of considerable importance to any simulation practitioner [31], there are not many papers that have contributed to methodologies of simulation software selection and simulation software evaluation techniques. Some literature concentrates on information about specific packages and a comparison of some of them (e.g. 11, 14, 15, 17), but it is beyond the scope of this paper to cover this, especially as the updating of packages changes previous comparisons [26].
A point which is usually discussed in related articles is simulation software classes. Currently, simulation software is classified into two major classes: languages and simulators 13, 27. Languages are either general purpose programming languages like FORTRAN, or general purpose simulation languages like GPSS, introduced in 1960 4, 12, 18, 23. A language is like a small foundry. You can and have to make any tool you need, but it takes time and needs expertise. A simulator, introduced in 1980 [4], is like a toolbox containing a limited number of different tools and maybe some flexible ones. The main advantage of using a simulator is that you do not need to spend time and effort on making tools, but the flexibility is not as great as the flexibility of a language. Because of the growth in quality, features and flexibility of simulators, they are also more than the languages used for simulation modelling these days. A survey by Hlupic and Paul [20]showed that less than 10% of the users at universities and industry use only simulation languages. The majority use either both simulators and simulation languages or just simulators (Table 1).
Programming-like commands and interfaces with programming languages are features which make a simulator flexible. In general, improvements in the facilities available in simulators makes them increasingly powerful, flexible and user friendly 23, 36. “The distinction between simulators and simulation languages is blurring. They are moving toward each other by offering special features” [2].
However, each class of simulation software has its own advantages 25, 26, 33, 38. For a historical background of simulation modelling software see Ref. [4]or [29].
Other classifications are introduced by some experts. Banks [2]counted spreadsheets and rapid modelling tools as two other classes of simulation modelling tools. Carson [8]classifies classes as: pure simulator, simulator with programming-like capability, simulation language with simulator-like extensions, and simulation languages.
Whatever the classification, what matters in simulation software selection is the capabilities, features and suitability for the specific application area rather than the way they are classified.
Although many of the tools are included in simulation software, other software such as animators, simulation support software 3, 41and input and output data analysers 12, 33, 38can be used in integration with the simulation software.
It should also be mentioned that whilst most literature surveyed covers discrete event simulation there are some others which concentrate on the evaluation of tools for continuous simulation [13].
Section snippets
Methodologies of simulation software selection
There are few attempts to develop a structured methodology for simulation software selection in the literature searched. A survey by Christy and Watson [9]showed how new programming languages are selected for simulation applications (Table 2).
A summary of what experts 2, 10, 19, 21, 25, 26have proposed as methodologies for simulation software selection is given in the following seven stages. Sometimes a stage consists of a numerous collection of tasks to be undertaken.
- 1.
Carry out a pre-selection
Simulation software evaluation techniques
Not much information can be found in the literature searched, regarding any techniques for the evaluation of simulation software.
Banks [2]suggested the following procedure for evaluation of packages resulting in a score for each package. First, assign a value between 0 and 10 to each factor (criteria). These would then be summed and normalised so that the total adds to a convenient number, say 100. He calls the resulting normalised number the factor weight. Then, each software would be scored
Criteria for simulation software selection
In contrast to the shortage of simulation software evaluation techniques and simulation software selection methodologies in the literature searched, many papers and books have stated their preferred list of important criteria for simulation software selection. Sometimes a phrase used by a paper or book for a criterion is different from the phrase used by another paper or book for the same criterion.
Usually a proposed list of criteria is classified into several groups (Table 4). Law and Haider
Guidelines and recommendations
The importance of simulation software selection can be realised from the fact that nearly all major books on simulation and introductory papers, beside the articles which discuss the matter directly, dedicate a section to the subject. Guidelines and recommendations relating to simulation software selection are given by the authors directly and indirectly. These can be found in simulation text books 5, 26, 35, introductory papers 12, 33, 38, 39, 41and papers mainly related to simulation software
Critique
Although a lot of the literature searched has given a list of criteria to be considered in simulation software selection, a lack of a standard common list is apparent. Experts have sometimes used terminology, for the criteria, which are not explained and are not understandable. The potential features of a package have not been presented in a way that can be updated and in a structure to be used for the test and evaluation of packages. Sample problems and case studies have been used to find out
Summary and conclusions
In this paper, the result of a search of the literature concerning simulation software selection has been discussed. Different levels of contributions of the literature are classified into four categories as: methodologies for simulation software selection, simulation software evaluation techniques, criteria to be used in the evaluation process, and recommendations on the subject. The lack of a proper methodology, a standard evaluation technique, a common list of criteria and a structured
References (42)
- et al.
Methodological approach to manufacturing simulation software selection
Computer Integrated Manufacturing
(1996) Interpreting simulation software checklists
ORMS Today
(1996)- J. Banks, Selecting simulation software, in: Proceedings of The 1991 Winter Simulation Conference, Arizona, USA, 1991,...
- J. Banks, Software for simulation, in: Proceedings of The 1995 Winter Simulation Conference, Virginia, USA, 1995, pp....
- et al.
The simulator: new member of the simulation family
Interfaces
(1991) - J. Banks, J.S. Carson, B.L. Nelson, Discrete-Event System Simulation, Prentice-Hall, Englewood Cliffs, NJ,...
- M. Bovone, V. De Ferrari, R. Manuelli, How to choose a useful simulation software, in: Proceedings of The 1989 European...
- et al.
Segmenting the simulation software market
OR Insight
(1990) - J. Carson, Simulation concepts in manufacturing and material handling, in: Autofact'90 Conference Proceedings, Detroit,...
- et al.
The application of simulation: a survey of industry practice
Interfaces
(1983)
Evaluation and selecting simulation software using the analytic hierarchy process
Integrated Manufacturing Systems
Manufacturing simulators: an evaluation and comparison
International Journal of Manufacturing System Design
Simulation software selection using SimSelect
Simulation
A critical evaluation of four manufacturing simulators
International Journal of Production Research
Simulation software in manufacturing environments: a users' survey
Journal of Computing and Information Technology – CIT
Selecting simulation software
OR Insight
Cited by (38)
Representing financial data streams in digital simulations to support data flow design for a future Digital Twin
2020, Robotics and Computer-Integrated ManufacturingCitation Excerpt :A significant volume of research has demonstrated the value of simulation to design and improve production systems [1–11].
Single-measure and multi-measure approach of predictive manufacturing control: A comparative study
2019, Computers and Industrial EngineeringCitation Excerpt :Effective decisions are taken only when the connection between different layers is taken into account. Simulation is a recommended technique that helps replicate real life scenarios while keeping different layers of the system intact (Nikoukaran & Paul, 1999; Thiede, Seow, Andersson, & Johansson, 2013). ( Mařík, Strasser, & Zoitl, 2009) defined simulation as “the dynamic representation of manufacturing facility by a computer model so that the impact of changes can be evaluated to support decision making process”.
An intelligent simulation environment for manufacturing systems
2014, Computers and Industrial EngineeringCitation Excerpt :This methodology is based on Business Process Modeling (BPM). The evaluated features are selected from a previous evaluation of Simulation Tools and a detailed list of collected features specified by Nikoukaran in Nikoukaran and Paul (1999). The evaluation detail includes features linked to each simulation phase proposed by Nikoukaran (Nikoukaran & Paul, 1999) and Banks, Carson, Nelson, and Nicol (2001).
Evaluating and selecting software packages: A review
2009, Information and Software TechnologyAn application of modeling and simulation to support industrial plants design
2012, International Journal of Modeling, Simulation, and Scientific ComputingDeveloping a prescriptive decision support system for shop floor control
2022, Industrial Management and Data Systems