Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


ucin1078512441.pdf (1.38 MB)

ETD Abstract Container

Abstract Header

ANALYSIS AND SENSITIVITY OF STOCHASTIC CAPACITATED MULTI-COMMODITY FLOWS

GHALEBSAZ-JEDDI, BABAK
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1078512441

Abstract Details

2004, MS, University of Cincinnati, Engineering : Industrial Engineering.
This research studies stochastic multi-commodity network flows which require specified demands to be satisfied at sink nodes. Link capacities fail with exponential rates and are returned to the network after being repaired after an exponentially distributed repair time. Systems under study are continuous time Markov chains. To our knowledge, sensitivity analysis of network flow solutions in terms of link capacities or demands, in deterministic or stochastic networks, has not been addressed in the network flow literature. We define sensitivity in terms of link criticality, and then utilize sensitivity information to estimate expected network availability in the long-run. The proposed methodology is simulation-based and combines deterministic network modeling with stochastic analysis. A linear programming problem that identifies feasible flows in a deterministic multi-commodity network with fixed capacities is formulated. This model is used within a simulation procedure to estimate the average network availability. Numerical examples illustrate the efficiency of the proposed methodologies for hot-spare and cold-spare networks. In the cold-spare network, the proposed methodology finds the route with minimum probability of failure anytime an event occurs. To increase the probability of system availability between any consecutive events, an improved model is suggested that finds a route that avoids critical links as much as possible. The research provides a methodology for investment analysis on link capacities. A criticality metric is proposed that identifies which link causes the majority of network failures. Adding one unit of capacity to this link is expected to have the largest increase in average network availability in the long-run.
Dr. Bruce Shultes (Advisor)
52 p.
capacitated network; multi-commodity; stochastic model; simulation; sensitivity analysis

Recommended Citations

Citations

  • GHALEBSAZ-JEDDI, B. (2004). ANALYSIS AND SENSITIVITY OF STOCHASTIC CAPACITATED MULTI-COMMODITY FLOWS [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1078512441

    APA Style (7th edition)

  • GHALEBSAZ-JEDDI, BABAK. ANALYSIS AND SENSITIVITY OF STOCHASTIC CAPACITATED MULTI-COMMODITY FLOWS. 2004. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1078512441.

    MLA Style (8th edition)

  • GHALEBSAZ-JEDDI, BABAK. "ANALYSIS AND SENSITIVITY OF STOCHASTIC CAPACITATED MULTI-COMMODITY FLOWS." Master's thesis, University of Cincinnati, 2004. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1078512441

    Chicago Manual of Style (17th edition)

ucin1078512441
609
© 2004, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.