Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 41 views

    Operations Research II: Stochastic Models Practice Final

    Uploaded by

    Eric Brugel
    1. The communicating classes of the Markov chain in question 1 are {1,2,3,4,7,8,10,11} and {5,6,9}. The states 1-11 are either positive recurrent or null recurrent with various periods. 2. The probability that at least two new jobs will arrive during the one hour repair time is approximately 0.48. 3. For the renewal process described, the probability of 2 events in 3 time units is 0.375, and the probability of more than 2 events between times 4 and 6 is 0.25.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    Operations Research II: Stochastic Models Practice Final

    Uploaded by

    Eric Brugel
    41 views2 pages
    1. The communicating classes of the Markov chain in question 1 are {1,2,3,4,7,8,10,11} and {5,6,9}. The states 1-11 are either positive recurrent or null recurrent with various periods. 2. The probability that at least two new jobs will arrive during the one hour repair time is approximately 0.48. 3. For the renewal process described, the probability of 2 events in 3 time units is 0.375, and the probability of more than 2 events between times 4 and 6 is 0.25.

    Original Description:

    stochastic models practice final

    Original Title

    Practice Final

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    1. The communicating classes of the Markov chain in question 1 are {1,2,3,4,7,8,10,11} and {5,6,9}. The states 1-11 are either positive recurrent or null recurrent with various periods. 2. The probability that at least two new jobs will arrive during the one hour repair time is approximately 0.48. 3. For the renewal process described, the probability of 2 events in 3 time units is 0.375, and the probability of more than 2 events between times 4 and 6 is 0.25.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    41 views2 pages

    Operations Research II: Stochastic Models Practice Final

    Uploaded by

    Eric Brugel
    1. The communicating classes of the Markov chain in question 1 are {1,2,3,4,7,8,10,11} and {5,6,9}. The states 1-11 are either positive recurrent or null recurrent with various periods. 2. The probability that at least two new jobs will arrive during the one hour repair time is approximately 0.48. 3. For the renewal process described, the probability of 2 events in 3 time units is 0.375, and the probability of more than 2 events between times 4 and 6 is 0.25.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    AMS 342

    Joe Mitchell

    Operations Research II: Stochastic Models Practice Final


    1. (10 points) Consider a discrete-time Markov Chain with transition diagram shown below. The state space is S = {1, 2, 3, . . .}, and an arc indicates a nonzero single-step transition probability.
    1 1 2 9 .1/2 3 4 7 8 10 .1/2 1/2 11 .1/2 1/2 0 .1/2 1/2

    5 6

    (a). (5 points) What are the communicating classes? (b). (5 points) For each state, determine if it is transient, positive recurrent, or null recurrent, and give its period. 2. (6 points) The jobs to be performed on a particular machine arrive according to a Poisson input process with a mean rate of three per hour. Suppose that the machine breaks down and will require (exactly) one hour to be repaired. What is the probability that at least two new jobs will arrive during the repair? 3. (6 points) Consider a renewal process in which the interevent times T i take on values 1, 2, or 3, with probabilities 1/2, 1/4, and 1/4, respectively. What is P {N (3) = 2}? What is P {N (4) N (6) > 2}? 4. (6 points) Consider an M/M/1 queue with arrival rate = 2 per hour and mean service time of 20 minutes. What is the probability that an average customer spends more than 20 minutes (total) in the system (including time in service)? 5. (6 points) Customers arrive to a bank with a single teller at rate 5 per hour. From statistical tests, we determine that the average time a customer waits in line is 27 minutes and that, on average, there are 3 customers (total, including anyone who might be in service) in the bank. What is the average length of time it takes the teller to serve a customer? 6. (12 points) Joe works in a busy AMS oce where students arrive according to a Poisson process with a mean interarrival time of 10 minutes. It takes Joe an amount of time X to serve a student, where X is uniformly distributed between 2 and 6 minutes. Immediately upon completion of a service, Joe takes a coee break, which lasts for a deterministic length of time of length 5 minutes. While Joe is serving a student or while he is on coee break, any arriving students to the oce turn around and go home. (a). (3 points) What fraction of time is Joe working to serve students? (b). (3 points) On the average, how many students per hour does Joe serve? (c). (3 points) What fraction of students that show up at the oce actually end up being served? (d). (3 points) Can this process be modelled as a CTMC? If so, how; if not, why not? 7. (8 points) Babies are born in Stony Brook Hospital in such a way that the time between the i th and (i + 1)st births has a distribution with density g (). The answers to the following questions may be left in terms of integrals. (a). (4 points) At what rate are babies born? (b). (4 points) Let Tlast be the time of the arrival of the last baby in the twentieth century, and let T f irst be the time of the arrival of the rst baby in the twenty-rst century. What is P {Tf irst Tlast > t}? 8. (7 points) Consider a queueing system in which customers arrive in cars according to a Poisson process, with one car arriving on average every 5 minutes. One third of all cars contain a single customer; two thirds contain 2 customers. There are two identical servers, each serving a customer in exponential time with an average service time of 2 minutes. Construct the rate transition diagram for the corresponding CTMC. Dene your states explicitly! 9. (8 points) Suppose that coin A has probability .4 of coming up heads, and that coin B has probability .5 of coming up heads. If the coin ipped today comes up heads, then we select coin A to ip tomorrow, and if it comes up tails, then we select coin B to ip tomorrow. Assume that the coin initially ipped (on day number 0) is equally likely to be coin A or coin B. We wish to model this process as a Markov chain. (a). (2 points) Specically, what is the state space, and what is the interpretation of being in state i at time n? (b). (2 points) What is the single-step transition matrix P? (Label your rows and columns with the states.)

    (c). (2 point) What is the probability that the third coin we ip is coin A? (That is, what is the probability that the coin that we ip on day number 2 is coin A?) (d). (2 point) What is the probability that the third coin we ip comes up heads? (That is, what is the probability that the coin that we ip on day number 2 comes up heads?) 10. (6 points) Consider a discrete-time Markov chain on state space S = {0, 1, 2} whose single-step probability transition matrix is given below. 1/2 0 1/2 P = 1/3 1/3 1/3 1/2 1/4 1/4 (a). (2 points) Compute P {X5 = 0 | X3 = 0}. (b). (4 points) Compute P {X2 = X3 | X0 = 0, X4 = 2}. 11. (5 points) Consider a CTMC on state space {0, 1, 2}. The values of the holding time parameters are 0 =??, 1 = 3, 2 =??. The Q matrix is given by ?? 1/2 1 Q = ?? ?? ?? 2 ?? 5 and the jump matrix is given by ?? P J = 1/5 ?? ?? ?? ?? ?? . ?? ??

    Fill in all of the blanks where I wrote ??. 12. (20 points) Consider the queueing network shown below, where each box (node) represents a queueing station with innite capacity (unlimited chairs). Assume that each node i of the network has a single server working at exponential rate i (1 = 12, 2 = 3, 3 = 4). Customers that leave node 1 go to node 3 with probability 1/4, while customers that leave node 2 depart the system with probability 1/2. There is a Poisson arrival stream of customers to node 1 with rate 2.

    1/2 Poisson rate 2 1 2 3

    1/4 (a). (4 points) Assume that the network is currently in the situation of having one customer at node 1, two customers at node 2, and three customers at node 3. What is the probability that the next change in state occurs within the next 4 units of time, but not within the next 2 units of time? (b). (2 points) What is the routing matrix R for the queueing network? (c). (4 points) What are the net eective arrival rates i at each node? (d). (5 points) What is the long-run probability that the network is in the situation of having one customer at node 1, two customers at node 2, and three customers at node 3? (e). (5 points) How would your answer to (d) change is there were an innite number of servers at node 3?

    You might also like