Computer Simulation in Management Science

by Pidd, Michael

• ISBN13: 9780470092309
• ISBN10: 0470092300
• Edition: 5th
• Format: Paperback
• Copyright: 2006-04-14
• Publisher: WILEY

Summary

The fifth edition of this book reflects its continued popularity and standing in the field. It provides a clear guide to the role of modelling and the computer simulation methods used in management science. Readers will find an in-depth coverage of the modelling, computing and statistical aspects of discrete simulation and systems dynamics.Overall the book shows how practical simulation models are built and used, and provides the theory needed to do this. Revisions to this edition include a new chapter on Monte Carlo simulation using spreadsheets, a new look inside discrete simulation software and simulation models in Visual Basic, SIMUL8 and Micro Saint.Further information can be found at:http://www.lancs.ac.uk/staff/smamp/mpsim.html

Author Biography

Michael Pidd is Professor of Management Science in the Management Science Department of Lancaster University Management School. He is author of Tools for Thinking: Modelling in Management Science, now in its second edition, and editor of Systems Modelling: Theory and Practice - both published by John Wiley & Sons. He is a consultant to businesses and the public sector, a former President of the Operational Research Society and is a Research Fellow in the UK's Advanced Institute for Management Research.

Table of Contents

Preface to the Fifth Edition	xv
Part I: Fundamentals Of Computer Simulation In Management Science	1	(60)
The computer simulation approach	3	(12)
Models, experiments and computers	3	(1)
Some applications of computer simulation	4	(3)
Manufacturing	4	(1)
Health care	5	(1)
Business process re-engineering	5	(1)
Transport systems	6	(1)
Defence	7	(1)
Models in management science	7	(1)
Simulation as experimentation	8	(1)
Why simulate?	9	(1)
Simulation versus direct experimentation	9	(1)
Simulation versus mathematical modelling	10	(1)
Summary	10	(5)
Exercises	11	(1)
References	11	(4)
A variety of modelling approaches	15	(14)
General considerations	15	(1)
Time handling	15	(3)
Time slicing	15	(2)
Next-event technique	17	(1)
Time slicing or next event?	18	(1)
Stochastic or deterministic?	18	(7)
Deterministic simulation: a time-slicing example	19	(2)
Stochastic simulation	21	(4)
Discrete or continuous change	25	(4)
Discrete change	26	(1)
Continuous change	26	(1)
A few words on simulation software	27	(1)
Exercises	27	(1)
References	28	(1)
Computer simulation in practice	29	(16)
Process, content, problem and project	29	(3)
Process and content	29	(1)
Problems and projects	30	(1)
Two parallel streams	30	(2)
The simulation problem part of the study	32	(1)
Problem structuring	32	(3)
3.3.1 Problem structuring as exploration	33	(2)
Modelling	35	(5)
Conceptual model building	35	(1)
Computer implementation	36	(1)
Validation	37	(1)
Experimentation	38	(1)
Implementation	39	(1)
The project part of the study	40	(5)
Initial negotiation and project definition	41	(1)
Project management and control	42	(1)
Project completion	43	(1)
Exercises	43	(1)
References	44	(1)
Static Monte Carlo simulation	45	(16)
Basic ideas	45	(4)
Risk and uncertainty	45	(1)
The replacement problem: a reprise	46	(1)
Static Monte Carlo simulation defined	47	(2)
Some important considerations	49	(2)
Subjective probabilities	49	(1)
Repeatability	50	(1)
Some simple static simulations	51	(6)
The loan repayment	51	(2)
An investment decision	53	(4)
Simulation on spreadsheets	57	(4)
Exercises	58	(1)
References	59	(2)
Part II: Discrete Event Simulation	61	(186)
Discrete event modelling	63	(20)
Fundamentals	63	(1)
Terminology	63	(3)
Objects of the system	64	(1)
The organization of entities	65	(1)
Operations of the entities	65	(1)
Activity cycle diagrams	66	(12)
Example 1: a simple job shop	68	(3)
Example 2: the harassed booking clerk	71	(3)
Example 3: the delivery depot	74	(3)
Using the activity cycle diagram	77	(1)
Activity cycle diagrams: a caveat	78	(5)
Exercises	79	(2)
References	81	(2)
How discrete simulation software works	83	(26)
Introduction	83	(2)
Why understand how simulation software is organized?	83	(1)
Simulation executives in more detail	84	(1)
Application logic	85	(1)
The three-phase approach	85	(5)
Bs	85	(1)
Cs	86	(1)
The exception to the general rule	87	(1)
Bs and Cs in the harassed booking clerk problem	88	(1)
Another example: a T-junction	89	(1)
How the three-phase approach works	90	(3)
The A phase	92	(1)
The B phase	92	(1)
The C phase	92	(1)
The harassed booking clerk---a manual three-phase simulation	93	(4)
The first A phase	94	(1)
The first B phase	94	(1)
The first C phase	95	(1)
The second A phase	95	(1)
The next B and C phases	96	(1)
The third A phase	96	(1)
The third B phase	96	(1)
The event-based worldview	97	(3)
Events in the harassed booking clerk problem	97	(2)
Event-based executives	99	(1)
The activity-scanning approach	100	(1)
Activities	100	(1)
Activity-scanning executives	101	(1)
Process-based approaches	101	(4)
Processes in the harassed booking clerk problem	102	(1)
Process interaction	103	(1)
Process-based executives	104	(1)
Which approach is best?	105	(4)
Three-phase versus process-based approaches	105	(1)
Exercises	106	(2)
References	108	(1)
Writing a three-phase simulation program	109	(20)
Introduction	109	(2)
The basic structure of the library	110	(1)
Inside the executive	111	(2)
The control array	112	(1)
Using the control array to operate a three-phase simulation	112	(1)
The Visual Basic implementation	113	(7)
Some comments on Visual Basic	113	(1)
The variables and their types	114	(2)
The A phase	116	(2)
The B phase	118	(1)
The C phase	119	(1)
Running the simulation	119	(1)
Using VBSim to simulate the harassed booking clerk problem	120	(6)
Entities, Bs and Cs	120	(1)
Personal enquirers and phone calls arrive	121	(2)
The end of personal service and phone calls	123	(1)
Observations	124	(1)
The Cs	124	(1)
Initialization and finalization	125	(1)
Putting it all together	126	(3)
Exercises	127	(1)
References	128	(1)
Visual interactive modelling and simulation	129	(20)
Basic ideas	129	(3)
Visual interactive modelling (VIM)	129	(1)
Visual simulation output	130	(1)
Interaction	131	(1)
A caveat	132	(1)
Designing a visual simulation display	132	(4)
Iconic displays	133	(1)
Logical displays	134	(1)
Chart displays	135	(1)
VIMS	136	(10)
Joe's exhaust parlour	136	(2)
Joe's exhaust parlour in Micro Saint: model building	138	(2)
Joe's exhaust parlour in Micro Saint: running and analysing the simulation	140	(3)
Joe's exhaust parlour in Simul8: model building	143	(2)
Joe's exhaust parlour in Simul8: running and analysing the simulation	145	(1)
Visual interactive simulation: a reprise	146	(3)
Exercises	146	(1)
References	146	(3)
Discrete simulation software	149	(26)
General principals	149	(1)
A quick overview of discrete simulation software	150	(1)
Vims and their relatives	151	(7)
Vims---a reprise	151	(1)
Block diagram systems	152	(5)
Vims and block diagram systems	157	(1)
Programming using a general purpose language	158	(3)
Pros and cons	158	(1)
Libraries and component-based software	159	(2)
Programming approaches using simulation languages	161	(7)
Common features of simulation languages	162	(1)
Anexample:Simscript Ii.5	163	(5)
Layered systems and application templates	168	(1)
Lyered systems	168	(1)
Application templates	169	(1)
Appraising simulation software: some principles	169	(3)
The type of application	170	(1)
The expectations for end use	170	(1)
Knowledge, computing policy and user support	171	(1)
Price	171	(1)
Which to choose? Horses for courses	172	(3)
Vims	172	(1)
Simulation languages	172	(1)
References	173	(2)
Sampling methods	175	(32)
Basic ideas	175	(4)
General principles of random sampling	176	(1)
Top-hat sampling	177	(1)
The fundamental random sampling process	178	(1)
Use of pre-written libraries of algorithms	179	(1)
Random number generation	179	(6)
Truly random numbers	179	(1)
Pseudo-random numbers	180	(1)
Congruential generators	181	(1)
General requirements for these generators	182	(1)
Multiplicative congruential generators	182	(2)
Improving on simple congruential generators	184	(1)
Using inbuilt random number generators	185	(1)
Testing random number generators	185	(4)
Scatter plots	186	(1)
Auxiliary sequences	187	(1)
Frequency tests	187	(1)
Serial test	188	(1)
Gap test	188	(1)
Other tests	188	(1)
General methods for random sampling from continuous distributions	189	(4)
Inversion	189	(3)
Rejection	192	(1)
Composition	193	(1)
Random sampling algorithms for discrete distributions	193	(4)
Sampling from histograms	193	(1)
Implicit inverse transformation	194	(2)
Discrete rejection---samples from a Poisson distribution	196	(1)
Sampling from the normal distribution	197	(4)
The original Box-Muller method	198	(1)
Box-Muller polar variation	199	(1)
Sampling from a normal distribution by composition	200	(1)
A poor way to sample from the normal distribution	201	(1)
Deriving one distribution from another---log-normal variates	201	(1)
Sampling from non-stationary processes: thinning	202	(5)
Exercises	203	(1)
References	204	(3)
Planning and analysing discrete simulation output	207	(26)
Fundamental ideas	207	(3)
Simulation as directed experimentation	207	(1)
Estimation and comparison	208	(1)
Three important principles	209	(1)
Some preliminary advice	210	(1)
Dealing with transient effects	210	(5)
Terminating and non-terminating systems	210	(2)
Achieving steady state	212	(1)
Using a run-in period	213	(1)
Welch's method for determining the run-in period	214	(1)
Dealing with lack of independence	215	(3)
Simple replication	215	(1)
Using batch means	216	(1)
Overlapping batch means (OBM)	217	(1)
Regenerative methods	218	(1)
Variance reduction	218	(8)
The basic problem---sampling variation	219	(1)
Set and sequence effects	220	(2)
Common random number streams and synchronization	222	(1)
Control variates (regression sampling)	223	(2)
Antithetic variates	225	(1)
Descriptive sampling	226	(2)
Basic idea	226	(1)
Procedure	227	(1)
Experimentation	228	(5)
Basic ideas	228	(1)
Factorial experiments	229	(2)
Exercises	231	(1)
References	231	(2)
Model Testing and Validation	233	(14)
The importance of validation	233	(4)
Validation is impossible, but desirable	233	(1)
Some practical issues	234	(1)
The ``real'' world, the model and observation	235	(1)
A The hypothetico-deductive approach	236	(1)
The importance of process and other aspects	237	(1)
Validation and comparison	237	(3)
Experimental frames	238	(1)
Program verification and model validation	239	(1)
Black box validation	240	(2)
Black box validation: a model's predictive power	240	(1)
How valid?	240	(1)
Validation errors	241	(1)
Testing model components	241	(1)
White box validation	242	(3)
Detailed internal structure	242	(1)
Input distributions	242	(1)
Static logic	243	(1)
Dynamic logic	244	(1)
Type zero errors	245	(2)
Over-elaboration	245	(1)
Over-simplification	246	(1)
Steering a sensible course	246	(1)
References	246	(1)
PART III: SYSTEM DYNAMICS	247	(50)
Structure, behaviour, events and Feedback systems	249	(14)
Events, behaviours and structures	249	(2)
System simulation	249	(1)
The importance of system structure	250	(1)
Feedback systems	251	(4)
Hierarchical feedback systems: an example	251	(2)
Causal loop diagrams	253	(2)
Modelling feedback systems	255	(3)
Delays	255	(1)
Levels and stocks	256	(1)
Rates and flows	257	(1)
Policies	258	(1)
The origings of system dynamics	258	(5)
Control theory	259	(1)
Exercises	260	(1)
References	261	(2)
System dynamics modelling and simulation	263	(16)
Introduction	263	(3)
Stock and flow diagrams	264	(1)
A stock and flow diagram for Big Al's problem	265	(1)
Beyond the diagrams---system dynamics simulation	266	(6)
Time handling in system dynamics	267	(1)
Equation types	268	(1)
Powersim equations for Big Al's problem	269	(1)
Integration and the value of dt	270	(2)
Simulating delays in system dynamics	272	(2)
Pipeline delays	272	(1)
Exponential delays	272	(2)
Information delays	274	(1)
System dynamics modelling	274	(5)
Modelling from the outside in	275	(1)
Modelling from the inside out	276	(1)
Exercises	277	(1)
References	277	(2)
System dynamics in practice	279	(18)
Associated Spares Ltd	279	(9)
The problem as originally posed	279	(1)
The multi-echelon system	280	(1)
The retail branch model	280	(3)
The regional warehouse model	283	(1)
The central warehouse model	284	(1)
The total system model	285	(1)
Some conclusions	286	(1)
A postscript	287	(1)
Dynastat Ltd	288	(5)
An expansion programme	288	(1)
The manpower problem	288	(1)
Recruitment	288	(1)
Turnover	289	(1)
Some effects of this structure	290	(1)
Validating the model	290	(1)
Simulation results	291	(1)
Predicting length of service	292	(1)
The value of the exercise to Dynastat	292	(1)
System dynamics in practice	293	(4)
Simple models	293	(1)
Communication	294	(1)
New thinking	294	(1)
Evolutionary involvement	295	(1)
References	295	(2)
Index	297