Geri Dön

Dynamic resource allocation in manufacturing and service industries

Başlık çevirisi mevcut değil.

  1. Tez No: 401057
  2. Yazar: TUBA YILMAZ
  3. Danışmanlar: PROF. DR. PINAR KESKİNOCAK, PROF. DR. ESMA GEL
  4. Tez Türü: Doktora
  5. Konular: Endüstri ve Endüstri Mühendisliği, Industrial and Industrial Engineering
  6. Anahtar Kelimeler: Belirtilmemiş.
  7. Yıl: 2013
  8. Dil: İngilizce
  9. Üniversite: Georgia Institute of Technology
  10. Enstitü: Yurtdışı Enstitü
  11. Ana Bilim Dalı: Belirtilmemiş.
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 131

Özet

Özet yok.

Özet (Çeviri)

In this thesis, we study three applications of dynamic resource allocation: the rst two consider dynamic lead-time quotation in make-to-order (MTO) systems with substitutable products and order cancellations, respectively; and the third application is a manpower allocation problem with job-teaming constraints. Matching supply and demand for manufacturing and service industries has been a fundamental focus of operations management literature, which concentrated on optimizing or improving supply-side decisions since demand has generally been as- sumed to be exogenously determined. However, recent business trends and advances in consumer behavior modeling have shown that demand for goods and services can clearly be shaped by various decisions that a rm makes, such as price and lead- time. In fact, competition between companies is no longer mainly based on price or product features; lead-time is one of the strategic measures to evaluate suppliers. In MTO manufacturing or service environments that aim to satisfy the customers' unique needs, lead-time quotation impacts the actual demand of the products and the overall pro tability of the rm. In the rst two parts of the thesis, we study the dynamic lead-time quotation problem in pure MTO (or service) systems characterized by lead-time sensitive Poisson demand and exponentially distributed service times. We formulate the problem as an in nite horizon Markov decision process (MDP) with the objective of maximizing the long-run expected average pro t per unit time, where pro ts are de ned to speci cally account for delays in delivery of the customer orders. We study dynamic lead-time quotation problem in two particular settings; one setting with the possibility of demand substitution and another setting with order cancellations. The fundamental trade-o in lead-time quotation is between quoting short lead-times and attaining them. In case of demand substitution, i.e., in presence of substitutable products and multiple customer classes with di erent requirements and margins, this trade-o also includes capacity allocation and order acceptance decisions. In particular, one needs to decide whether to allocate capacity to a low- margin order now, or whether to reserve capacity for potential future arrivals of high-margin orders by considering customer preferences, the current workload in the system, and the future arrivals. In the case of order cancellations, one needs to take into account the probability of cancellation of orders currently in the system and quote lead-times accordingly; otherwise quotation of a longer lead-time may result in the loss of customer order, lower utilization of resources, and, in turn, reduced in pro ts. In Chapter 2, we study a dynamic lead-time quotation problem in a MTO system with two (partially) substitutable products and two classes of customers. Customers decide to place an order on one of the products or not to place an order, based on the quoted lead-times. We analyze the optimal pro t and the structure of the optimal lead-time policy. We also compare the lead-time quotes and pro ts for di erent quota- tion strategies (static vs. dynamic) with or without substitution. Numerical results show that substitution and dynamic quotation have synergetic e ects, and higher bene ts can be obtained by dynamic quotation and/or substitution when di erence in product revenues or arrival rates, or total trac intensity are higher. In Chapter 3, we study a dynamic lead-time quotation problem in a MTO sys- tem with single product considering the order cancellations. The order cancellations can take place during the period that the order is being processed (either waiting or undergoing processing), or after the processing is completed, at the delivery to the customer. We analyze the behavior of optimal pro t in terms of cancellation param- eters. We show that the optimal pro t does not necessarily decrease as cancellation rate increases through a numerical study. When the pro t from a cancelled order, xii arrival rate of customers, or lead-time sensitivity of customers are high, there is a higher probability that optimal pro t increases as cancellation rate increases. We also compare the cancellation scenarios with the corresponding no-cancellation sce- narios, and show that there exists a cancellation scenario that is at least as good in terms of pro t than a no-cancellation scenario for most of the parameter settings. In Chapter 4, we study the Manpower Allocation Problem with Job-Teaming Con- straints with the objective of minimizing the total completion time of all tasks. The problem arises in various contexts where tasks require cooperation between workers: a team of individuals with varied expertise required in di erent locations in a business environment, surgeries requiring di erent composition of doctors and nurses in a hos- pital, a combination of technicians with individual skills needed in a service company. A set of tasks at random locations require a set of capabilities to be accomplished, and workers have unique capabilities that are required by several tasks. Tasks require synchronization of workers to be accomplished, hence workers arriving early at a task have to wait for other required workers to arrive in order to start processing. We present a mixed integer programming formulation, strengthen it by adding cuts and propose heuristic approaches. Experimental results are reported for low and high coordination levels, i.e., number of workers that are required to work simultaneously on a given task.

Benzer Tezler

  1. Proje yönetiminde kantitatif yöntemlerin uygulanması

    The application of the quantitative methods in project management

    ZİYA ULUKAN

  2. Atölyede iş çizelgeme

    Operations scheduling in job shops

    GÖKHAN KIPÇAK

    Yüksek Lisans

    Türkçe

    Türkçe

    1990

    Endüstri ve Endüstri Mühendisliğiİstanbul Teknik Üniversitesi

    PROF.DR. ATAÇ SOYSAL

  3. Gelişmekte olan ülkelerde teknoloji politikalarının belirlenmesi ve Türkiye'deki durum

    Technology policies in developing countries and the situatiın in Turkey

    MEHPARE BARIŞ

    Yüksek Lisans

    Türkçe

    Türkçe

    1997

    Endüstri ve Endüstri Mühendisliğiİstanbul Teknik Üniversitesi

    Mühendislik Yönetimi Ana Bilim Dalı

    DOÇ. DR. TUFAN V. KOÇ

  4. Sağlık sistemlerinde çok amaçlı karar verme

    Multiple creteria decision making in health systems

    YAŞAR SAADETTİN PEKTUT

    Yüksek Lisans

    Türkçe

    Türkçe

    1990

    Endüstri ve Endüstri Mühendisliğiİstanbul Teknik Üniversitesi

    Y.DOÇ.DR. MEHMET TANYAŞ

  5. Stratejik kalite planlaması

    Strategic quality planning

    AYNUR ÖZYILMAZ

    Yüksek Lisans

    Türkçe

    Türkçe

    1998

    Endüstri ve Endüstri Mühendisliğiİstanbul Teknik Üniversitesi

    Endüstri Mühendisliği Ana Bilim Dalı

    PROF. DR. M. BÜLENT DURMUŞOĞLU