models to estimate arrival counts and staffing requirements in nonstationary queueing systems ...

137 Pages·2011·2.23 MB·English

by  Fairweather, Lindon

MODELS TO ESTIMATE ARRIVAL COUNTS AND STAFFING
REQUIREMENTS IN NONSTATIONARY QUEUEING SYSTEMS APPLIED TO LONG
DISTANCE ROAD RACES

by

LINDON P. FAIRWEATHER
B.A., Eastern Connecticut State University, 1999
M.A., University of Connecticut, 2002
M.S., University of Central Florida, 2005

A dissertation submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
in the department of Industrial Engineering and Management Systems
in the College of Engineering and Computer Science
University of Central Florida
Orlando, Florida

Summer Term
2011

Major professor: Charles H. Reilly
ABSTRACT

We examine the problem of staffing refreshment stations at a long distance road race. A
race is modeled as a mixed queueing network in which the required number of servers at each
service station has to be estimated. Two models to represent the progress of runners along a long
distance road race course are developed. One model is a single-class model that allows a road
race manager to staff service stations assuming the runners are identical to those in some
historical dataset. Another model is a multi-class simulation model that allows a road race
manager to simulate a race of any number of runners, classified based on their running pace into
different runner classes.
Both the single-class model and the multi-class model include estimates for the
rates at which the runners arrive at specified locations along the course. The arrival rates,
combined with assumed service rates, allow us to base staffing decisions on the Erlang loss
formula or a lesser known staffing rule that gives a lower bound for the required number of
servers. We develop a staffing strategy that we call the Peak Arrival Staffing Bound (PASB),
which is based on this staffing bound.
The PASB and the Erlang loss formula are implemented in the single-class model
and the multi-class simulation model. By way of numerical experiments, we find that the PASB
is numerically stable and can be used to get staffing results regardless of the traffic intensity.
This finding is in contrast to the Erlang loss formula, which is known to become numerically
unstable and overflows when the traffic intensity exceeds 171. We compare numerical results of
the PASB and the Erlang loss formula with a blocking probability level of 5% and find that when

ii
the traffic intensity is high, staffing results based on the PASB are more conservative than
staffing results based on the Erlang loss formula. As the traffic intensity gets lower, we find that
staffing results based on the PASB are similar to staffing results based on the Erlang loss
formula. These findings suggest that the PASB can be a valuable tool to aid race directors in
making staffing decisions for races of all traffic intensities.

iii
To Bubbles, Peppy, and Vicky

iv
TABLE OF CONTENTS
LIST OF FIGURES ....................................................................................................................... ix
LIST OF TABLES .......................................................................................................................... x
CHAPTER 1 INTRODUCTION TO ROAD RACE MODELING ............................................... 1
1.1 Motivating Problem.......................................................................................................... 1
1.2 Problem Description ......................................................................................................... 1
1.3 The Research Plan ............................................................................................................ 3
CHAPTER 2 LITERATURE REVIEW ......................................................................................... 5
2.1 Introduction ......................................................................................................................... 5
2.2 Road Race Modeling ........................................................................................................... 5
2.3 Queueing Networks ............................................................................................................. 6
2.3.1 Queueing Theory Nomenclature .................................................................................. 7
2.3.2 Examples of Queueing Network Models ..................................................................... 9
2.4 Erlang Loss Formula ......................................................................................................... 19
2.4.1 The Simple Stationary Approximation (SSA) Method .............................................. 21
2.4.2 The Pointwise Stationary Approximation (PSA) Method .......................................... 21
2.4.3 The Simple Peak Hour Approximation (SPHA) Method ........................................... 22
2.4.4 The Fixed Point Approximation (FPA) Method ........................................................ 23
2.5 Simulation and Monte Carlo Methods .............................................................................. 23

v
2.5.1 Simulation................................................................................................................... 24
2.5.2 Monte Carlo Methods ................................................................................................. 24
2.6 Some Applications of Monte Carlo Methods and Simulation .......................................... 27
2.6.1 Monte Carlo Methods and Simulation: Biochemistry ................................................ 27
2.6.2 Monte Carlo Methods and Simulation: Environmental and Water Engineering ....... 27
2.6.3 Monte Carlo Methods and Simulation: Maritime Science ......................................... 28
2.7 Variance Reduction Techniques ........................................................................................ 29
2.7.1 Importance Sampling.................................................................................................. 29
2.7.2 Stratified Sampling ..................................................................................................... 30
2.8 Chapter Summary .............................................................................................................. 30
CHAPTER 3 MODEL DEVELOPMENT AND METHODOLOGY .......................................... 32
3.1 Introduction ....................................................................................................................... 32
3.2 A Model of a Road Race Traffic System Based on Available Historical Data................. 33
3.2.1 Estimating the Pace of a Runner ................................................................................ 34
3.2.2 Estimating Running Times ......................................................................................... 34
3.2.3 Estimating the Arrival Rates ...................................................................................... 36
3.2.4 Procedure 1: A Procedure to Estimate Peak Arrival Rates from Historical Data ...... 38
3.3 A Framework to Estimate Staffing Requirements in a Long Distance Road Race ........... 41
3.3.1 Background................................................................................................................. 41

vi
3.3.2 A Staffing Equation Based on Peak Arrival Rates ..................................................... 42
3.4 Chapter Summary .............................................................................................................. 43
CHAPTER 4 EXPERIMENTS AND RESULTS ......................................................................... 44
4.1 Introduction ....................................................................................................................... 44
4.2 Data Description and Summary Statistics ......................................................................... 44
4.3 Estimating the Peak Arrival Rates .................................................................................... 45
4.4 Experiment A: Estimating Staffing Requirements ............................................................ 49
4.4.1 Experiment A-1: Staffing with Erlang Loss Formula ................................................ 49
4.4.2 Experiment A-2: Staffing with the Peak Arrival Staffing Bound .............................. 54
4.4.3 Experiment B .............................................................................................................. 59
4.4.4 Experiment B-1: Staffing with Erlang Loss Formula................................................. 63
4.4.5 Experiment B-2: Staffing with the Peak Arrival Staffing Bound ............................... 67
4.5 Additional Experiments ..................................................................................................... 69
4.5.1 Experiments ................................................................................................................ 69
4.5.2 Rationale for Choice of Experiments ......................................................................... 71
4.6 Analysis of Results ............................................................................................................ 87
4.7 Chapter Summary .............................................................................................................. 88
CHAPTER 5 A MULTI-CLASS MONTE CARLO SIMULATION MODEL OF A LONG
DISTANCE ROAD RACE ........................................................................................................... 90

vii
5.1 Simulation Model Development ....................................................................................... 90
5.1.1 The Multiclass Simulation Model .............................................................................. 91
5.1.2 Implementing the Multiclass Simulation Model ........................................................ 93
5.1.3 Alternative Approaches and Implementation Limitations ....................................... 100
5.2 Experiments with the Simulation Model ......................................................................... 101
5.2.1 Runner-Mix Experiments ......................................................................................... 102
5.2.2 Race Size Experiment............................................................................................... 112
5.3 Chapter Summary ............................................................................................................ 116
CHAPTER 6 CONCLUSIONS AND FUTURE RESEARCH ................................................. 118
6.1 Contribution Summary .................................................................................................... 118
6.2 Recommendations and Directions for Future Research .................................................. 121
REFERENCES ........................................................................................................................... 122

viii
LIST OF FIGURES
Figure 1: Staffing Based on Erlang Loss formula (Screenshot for Mile 1) .................................. 51
Figure 2: Number of Servers for Experiments 1 and 5 ................................................................. 76
Figure 3: Number of Servers for Experiments 2 and 6 ................................................................. 79
Figure 4: Number of Servers for Experiment 3 and 7 .................................................................. 82
Figure 5: Number of Servers for Experiments 4 and 8 ................................................................. 85
Figure 6: Simulation Implementation: Steps 0 – 1 ....................................................................... 95
Figure 7: Simulation Implementation: Steps 2 and 3.................................................................... 97
Figure 8: Simulation Implementation: Steps 2 and 3.................................................................... 98
Figure 9: Simulation Implementation: Steps 2 and 3.................................................................... 99

ix
LIST OF TABLES
Table 3.1: Time-Counts Table ...................................................................................................... 39
Table 2: Peak Arrival Rates and Time of Peak Arrivals ............................................................... 39
Table 3: Arrival Counts for Miles 3, 7, and 11 ............................................................................. 45
Table 4: Peak Arrival Rates and Traffic Intensity Estimates ........................................................ 46
Table 5: Staffing Estimates for Experiment A-1 .......................................................................... 52
Table 6: Arrival Rates and Staffing Estimates for Experiment A ................................................ 56
Table 7: Peak Arrival Rates and Traffic Intensity Estimates ........................................................ 60
Table 8: Arrival Rates, Traffic Intensity Estimates, and .............................................................. 64
Table 9: Arrival Rates, Traffic Intensity Estimates, and Staffing................................................. 67
Table 10: List of Experiments....................................................................................................... 70
Table 11: Number of Servers for Experiments 1 and 5 ................................................................ 72
Table 12: Number of Servers for Experiments 2 and 6 ................................................................ 77
Table 13: Number of Servers for Experiment 3 and 7 .................................................................. 80
Table 14: Number of Servers for Experiments 4 and 8 ................................................................ 83
Table 15: Class and Pace Distribution Table ................................................................................ 91
Table 16: Pace Distribution and Probability of Occurrence ......................................................... 93
Table 17: Runners Pace Distribution and Probability of Occurrence ......................................... 103
Table 18: Server Requirements Based on 10 Monte Carlo Replications of Experiment 5.1 Using
the PASB with 12 ................................................................................................................. 105
Table 19: Server Requirements Based on 10 Monte Carlo Replications of Experiment 5.1 Using
Erlang Loss Formula with Blocking Probability <=0.05 and 12 ......................................... 107

x