1、附录 A 英文资料 Simulation Modeling and Analysis The nature of simulation This is a paper about techniques for using computers to imitate, or simulate, the operations ofvarious kinds of real-world facilities or processes. The facility or process of interest is usuallycalled a system, and in order to study
2、 it scientifically we often have to make a set of assumptionsabout how it works. These assumptions, which usually take the form of mathematical or logicalrelationships, constitute a model that is used to try to gain some understanding of how thecorresponding system behaves. If the relationships that
3、 compose the model are simple enough, it may be possible to usemathematical methods (such as algebra, calculus, or probability theory) to obtain exact informationon questions of interest； this is called an analytic solution. However, most real-world systems aretoo complex to allow realistic models t
4、o be evaluated analytically, and these models must be studiedby means of simulation. In a simulation we use a computer to evaluate a model numerically, anddata are gathered in order to estimate the desired true characteristics of the model. As an example of the use of simulation, consider a manufact
5、uring company that is contemplatingbuilding a large extension onto one of its plants but is not sure if the potential gain in productivitywould justify the construction cost. It certainly would not be cost-effective to build the extensionand then remove it later if it does not work out. However, a c
6、areful simulation study could shedsome light on the question by simulating the operation of the plant as it currently exists and as itwould be if the plant were expanded. Application areas for simulation are numerous and diverse. Below is a list of some payof problems for which simulation has been f
7、ound to be a useful and powerful tool: Designing and analyzing manufacturing systems Evaluating military weapons systems or their logistics requirements Determining hardware requirements or protocols for communications networks Determining hardware and software requirements for a computer system Des
8、igning and operating transportation systems such as airports,freeways,ports,and subways Evaluating designs for service organizations such ascall centers, fast-food restaurantshospitals, and post offices Reengineering of business processes Determining ordering policies for an inventory Analyzing fina
9、ncial or economic systems Simulation is one of the most widely used operations research and management-sciencetechniques, if not the most widely used. One indication of this is the Winter SimulationConference, which attracts 600 to 700 people every year. In addition, there are several simulationvend
10、or users conferences with mor e than 100 participants per year. There are also several surveys related to the use of operations research techniques. For example,Lane, Mansour, and Harpell ( 1993 ) reported from a longitudinal study, spanning 1973 through1988, that simulation was consistently ranked
11、as one of the three most important operations-research techniques. The other two were math programming (a catch-all term that includesmany individual techniques such as linear programming, nonlinear programming, etc.) andstatistics (which is not an operations-research technique per se). Gupta (1997)
12、 analyzed 1294papers from the journal Interfaces (one of the leading journals dealing with applications ofoperations research) from 1970 through 1992, and found that simulation was second only tomath programming among 13 techniques considered. There have been, however, several impediments to even wi
13、der acceptance and usefulness ofsimulation. First, models used to study large-scale systems tend to be very complex, and writingcomputer programs to execute them can be an arduous task indeed. This task has been made mucheasier in recent years by the development of excellent software products that a
14、utomatically providemany of the features needed to program a simulation model. A second problem with simulationof complex systems is that a large amount of computer time is sometimes required. However, thisdifficulty is becoming much less severe as computers become faster and cheaper. Finally, there
15、appears to be an unfortunate impression that simulation is just an exercise in computerprogramming, albeit a complicated one. Consequently, many simulation studies have beencomposed of heuristic model building, coding, and a single run of the program to obtain theanswer. We fear that this attitude,
16、which neglects the important issue of how a properly codedmodel should be used to make inferences about the system of interest, has doubtless led toerroneous conclusions being drawn from many simulation studies. Systems, models, and simulation A system is defined to be a collection of entities, e.g.
17、 people or machines that act and interacttogether toward the accomplishment of some logical end. In practice, what is meant by thesystem depends on the objectives of a particular study. The collection of entities that comprise asystem for one study might be only a subset of the overall system for an
18、other. For example, if onewants to study a bank to determine the number of tellers needed to provide adequate service forcustomers who want just to cash a check or make a savings deposit, the system can be defined to bethat portion of the bank consisting of the tellers and the customers waiting in l
19、ine or being served.If, on the other hand, the loan officer and the safety deposit boxes are to be included, the definitionof the system must be expanded in an obvious way. We define the state of a system to be thatcollection of variables necessary to describe a system at a particular time, relative
20、 to the objectivesof a study. In a study of a bank, examples of possible state variables are the number of busy tellers, the number of customers in the bank, and the time of arrival of each customer in the bank. We categorize systems to be of two types, discrete and continuous. A discrete system is
21、one forwhich the state variables change instantaneously at separated points in time. A bank is an exampleof a discrete system, since state variables - e. g. the number of customers in the bank changeonly when a customer arrives or when a customer finishes being served and departs. A continuoussystem
22、 is one for which the state variables change continuously with respect to time. An airplanemoving through the air is an example of a continuous system, since state variables such as positionand velocity can change continuously with respect to time. Few systems in practice are whollydiscrete or wholl
23、y continuous； but since one type of change predominates for most systems, it willusually be possible to classify a system as being either discrete or continuous. At some points in the lives of most systems, there is a need to study them to try to gain someinsightinto the relationships among various
24、components, or to predict performance under some new conditions. Figure 1. Ways to study a system Experiment with the Actual System vs. Experiment with a Model of the System. If it ispossible ( and cost-effective) to alter the system physically and then let it operate under thenew conditions, it is
25、probably desirable to do so, for in this case there is no question about whether what we study is valid. However, it is rarely feasible to do this, because such an experiment would often be too costly or too disruptive to the system. For example, a bank may be contemplating reducing the number of te
26、llers to decrease costs, but actually trying this could lead to long customer delays and alienation. More graphically, the system might not even exist, but we nevertheless want to study it in its various proposed alternative configurations to see how it should be built in the first place； examples o
27、 f this situation might be a proposed communications network, or a strategic nuclear weapons system. For these reasons, it is usually necessary to build a model as a representation of the system and study it as a surrogate for the actual system. When using a model, there is always the question of wh
28、ether it accurately reflects the system for the purposes of the decisions to be made. Physical Model vs. Mathematical Model. To most people, the word model evokes images of clay cars in wind tunnels, cockpits disconnected from their air planes to be used in pilot training, or miniature supertankers
29、scurrying about in a swimming pool. These are examples of physical models (also called iconic models), and are not typical of the kinds of models that are usually of interest in operations research and systems analysis. Occasionally, however, it has been found useful to build physical models to stud
30、y engineering or management systems； examples include tabletop scale models of material-handling systems, and in at least one case a full-scale physical model of a fast- food restaurant inside a warehouse, complete with full-scale, real (and presumably hungry) humans. But the vast majority of models
31、 built for such purposes are mathematical, representing a system in terms of logical and quantitative relationships that are then manipulated and changed to see how the model reacts, and thus how the system would react-if the mathematical model is a valid one. Perhaps the simplest example of a mathematical model is the
