McMaster University

DeGroote School of Business

Fall 2010
Bus Q771: Stochastic Processes with Business Applications
Dr. Mahmut Parlar

ASSIGNMENT #1

Due date: Four working days before the midterm exam. For example, if the exam is to take
place on October 21, the due date would be October 17, at 4:00 p.m.

Each question is assigned 5 marks. (Maximum mark = 5 15 = 75.)

1 Review of Basic Probability Theory

1. (Bu¤ on’s Needle Problem) Suppose a ‡oor is marked with parallel lines all at a distance
2a apart. A needle of length 2` (` < a) is tossed at random on the ‡oor. Determine the
probability that the needle will intersect one of the lines.
Hint: Let X be the distance from centre of the needle to the nearest line, and Y be the angle
between the needle and direction perpendicular to the given lines. Here, “at random”means
that the distribution of X is uniform over [0; a] and Y is uniform over [ =2; =2], i.e.,

fX (x) = 1=a; 0 x a;
fY ( ) = 1= ; =2 =2:

2. Ch. 1, Pr. 3 (p. 37) in Kao.

3. Ch. 1, Pr. 15 (p. 39) in Kao.

4. Ch. 1, Pr. 16 (p. 39) in Kao.

5. Ch. 1, Pr. 35 (p. 42) in Kao.

1
2 Exponential Distribution and the Poisson Process
6. Consider the computer system example discussed in the class. The di¤erential equation
system for this M (t)=M=1=4 queue is
2 3
(t) (t)
6 7
6 [ (t) + ] (t) 7
6 7
[P0 (t); : : : ; P4 (t)] = [P0 (t); : : : ; P4 (t)] 6
0 0
6 [ (t) + ] (t) 7
7
6 7
4 [ (t) + ] (t) 5

where

t t t t
(t) = 8:924 1:584 cos + 7:897 sin 10:434 cos + 4:293 cos
1:51 3:02 4:53 6:04

and = 7 units/hour. Compute numerically the solution of this system and comment on the
transient probability of loss at time t, i.e., P4 (t).

7. Customers arrive at a facility according to a Poisson process of rate . Each customer pays
$1 on arrival, and it is desired to evaluate the expected value of the total sum collected during
the interval (0; t] discounted back to time 0. This quantity is given by
0 1
N (t)
X
M =E@ e Wk A

k=1

where is the discount rate, W1 ; W2 ; : : : are the arrival times and N (t) is the total number
of arrivals in (0; t].

(a) Compute M .
(b) What happens to M as t ! 1? Is the result intuitive?

8. Ch. 2, Pr. 2 (p. 87) in Kao.

9. Ch. 2, Pr. 6 (p. 88) in Kao.

10. Ch. 2, Pr. 14 (p. 90) in Kao.

3 Renewal Theory
11. The lifetime of a car has a distribution H and density h. Ms. Jones buys a new car as soon
as her old car either breaks down or reaches the age of T years. A new car costs $C1 and

2
 additional cost of $C2 is incurred whenever a car breaks down.

(a) Assuming that a T year old car in working order has an expected resale value R(T ),
what is Ms. Jones’long-run average cost?
(b) If H is uniform over (2; 8) and C1 = 4, C2 = 1, and R(T ) = 4 T =2, then what value of
T minimizes Ms. Jones’objective function?

12. Suppose a periodic review inventory system for which the demands X1 ; X2 ; : : : for a single
product in the successive weeks 1; 2; : : : are i.i.d. r.v.’s with density f (x) with …nite …rst
two moments. Any demand exceeding the current inventory is backlogged until inventory
becomes available by the arrival of a replenishment order. The inventory position is reviewed
at the beginning of each week and is controlled by an (s; S) rule with 0 s < S. Under this
control rule a replenishment order of size S x is placed when at a review the inventory level
x is at or below the reorder point s; otherwise no ordering is done. Assume instantaneous
delivery of every replenishment order.
Find the average frequency of ordering and the average order size.
Hint: Assume that S s is su¢ ciently large relative to the average weekly demand and use
the relevant asymptotic results.

13. A conceptual model of how telephone users behave after getting a busy signal, which is
consistent with observed behaviour, is the following. At time zero, a call is placed and a busy
signal is heard. The caller then ‡ips a coin that has probability p of landing heads. If the
coin lands tails, no further calls are made. If it lands heads, the caller chooses a random time
X1 , with F (x) = Pr(X1 x), and another call is placed at X1 . This call is successful with
probability 1 r. If another busy signal is heard, the process repeats itself.
Each call other than the original call placed at time zero is called a reattempt. In designing
telephone systems, often it is important to know the statistical behaviour of the reattempts.

(a) Find R(t) = mean number of reattempts made by t.

Hint: Condition on X1 = x and then use Laplace transforms.

(b) Assume F (t) = 1 e t and …nd the exact expression for R(t).

14. Ch. 3, Pr. 2 (p. 146) in Kao.

15. Ch. 3, Pr. 3 (p. 146) in Kao. [Typo in text; f (x) is f (t).]