Geri Dön

Deprem sonrası için geliştirilen ulaşım talep modeli: Elazığ örneği

A transportation demand model developed for the post-earthquake period: The case of Elaziğ

PDF İndir

  1. Tez No: 1019321
  2. Yazar: AYŞE POLAT
  3. Danışmanlar: DOÇ. DR. HÜSEYİN ONUR TEZCAN
  4. Tez Türü: Doktora
  5. Konular: Ulaşım, Transportation
  6. Anahtar Kelimeler: Belirtilmemiş.
  7. Yıl: 2026
  8. Dil: Türkçe
  9. Üniversite: İstanbul Teknik Üniversitesi
  10. Enstitü: Lisansüstü Eğitim Enstitüsü
  11. Ana Bilim Dalı: İnşaat Mühendisliği Ana Bilim Dalı
  12. Bilim Dalı: Ulaştırma Mühendisliği Bilim Dalı
  13. Sayfa Sayısı: Belirtilmemiş.

Özet

Deprem sonrası arama-kurtarma çalışmalarının ve yardım faaliyetlerinin sağlıklı yürütülebilmesi için, yollar veya ulaşım sistemlerinde oluşabilecek hasarların yanında, bireylerin ani ulaşım talebi sonucu oluşan trafik tıkanıklıkları da büyük bir sorun oluşturmaktadır. Bu nedenle, deprem sonrası yolculuk taleplerinin ayrıntılı bir şekilde incelenmesi afet yönetimi için önemlidir. Bu kapsamda, bu çalışmada 24 Ocak 2020 tarihinde tsunami tehlikesi barındırmayan bir bölge olan Elazığ'da meydana gelen 6,8 büyüklüğündeki deprem sonrası için, kent ulaşımı geliştirilen dört aşamalı ulaşım modeli kullanılarak modellenmiştir. Çalışmada, Elâzığ kent merkezinde yaşayan 2.739 depremzedenin katılımıyla gerçekleştirilen anket verileri kullanılmıştır. Analiz sonuçları, katılımcıların %75'inin depremin sonrasındaki ilk 24 saat içerisinde en az bir yolculuk yaptığını ve yolculukların deprem anına yakın zaman dilimlerinde yoğunlaştığını göstermektedir. Tez kapsamında kullanılan modelde, klasik dört aşamalı ulaşım modelinin yolculuk üretimi, yolculuk dağıtımı ve tür seçimi aşamaları uygulanırken, trafik ataması aşamasında istek hatları hazırlanmıştır. Bununla birlikte, dört aşamalı ulaşım modelinde yöntemsel olarak bazı değişiklikler yapılarak model deprem sonrası için düzenlenmiştir. Yapılan değişikliklerden biri, modelin yolculuk üretim ve çekim aşamasında yolculukların tahmininde lojit modelin kullanılmasıdır. Bu tercihin öncelikli nedeni, lojit modelin nicel olarak ifade edilemeyen soyut parametreleri de modelin içerisinde yer alan rassal bölümünde barındırmasından kaynaklanmıştır. Depremler gibi olağanüstü olaylar için ortaya çıkabilecek yolculuk talebinin, yalnızca sosyo-ekonomik değişkenlerin doğrusal bir fonksiyonu ile açıklanması doğru olmayacaktır. Bireylerin algıları, kısıtları ve karar süreçlerinin bu olağanüstü durumların modellenmesinde, ihmal edilebilecek özelliklerin olmadığı düşünülmektedir. Yapılan diğer bir değişiklik ise yolculuk dağıtım aşamasında kullanılan yerçekimi modelinde, kalibrasyon yapılmaması bunun yerine modelde düzenlemeler yapılıp, lojit model kullanılarak, yolculuk çekim olasılıklarının belirlenmesidir. Tür seçimi modeli ise klasik dört aşamalı ulaşım modeli ile aynı şekilde ve lojit model kullanılarak kurgulanmış bununla birlikte bağımsız değişkenler içerisinde depremle ilgili değişkenlere de yer verilmiştir. Trafik ataması aşamasında ise deprem sonrasında ortaya çıkabilecek birçok belirsizlik olduğu için istek hatlarının kullanılmasının daha doğru olacağı düşünülmüştür. Modelin tüm aşamalarında yer alan lojit modellerde hasarlı bina oranı değişkeni istatistiksel olarak anlamlı değerler almıştır. Bu doğrultuda, bu değişkenin modelin tüm aşamalarında belirleyici bir değişken olduğu sonucuna ulaşılmıştır. Bununla birlikte, TAB bazlı değerlendirmelerde, özellikle Ataşehir, Cumhuriyet ve Sürsürü mahallelerinin TAB içerisinde ve Çaydaçıra mahallesinde ise diğer TAB'lardan çok fazla yolculuk çekimi gerçekleştiği görülmüştür. Bu mahallelerin ulaşım güzergâhlarının, deprem sonrası ulaşım açısından kentte kritik bölgeler olduğu tespit edilmiştir. Bu tez çalışması, ülkemizde deprem sonrası ani yolculuk talebini, dört aşamalı ulaşım modeli kapsamında ele alan ilk çalışmalardan biridir. Geliştirilen model, afetlere hazırlık sürecinde ulaşım planlamasına davranış temelli bir bakış açısı sunmaktadır. Elde edilen sonuçların, karar vericilerin deprem sonrası ulaşımı daha etkin yönetebilmesine ve bu doğrultuda politika ve stratejiler geliştirebilmesine katkı sağlaması beklenmektedir.

Özet (Çeviri)

Earthquakes are among the most frequent natural disasters in Türkiye, causing significant loss of life and property. In addition to the direct effects of earthquakes, such as physical destruction, there are also indirect effects, such as disruptions to transportation infrastructure, that affect the functioning of critical systems. Moreover, both possible damage to roads and transportation systems during post-earthquake response efforts and traffic congestion due to a sudden increase in travel demand are major issues. Since earthquakes occur without warning, people tend to make trips for different purposes, such as gathering, evacuation, and other needs, at the same time. For this reason, a detailed analysis of post-earthquake travel demand is essential for effective disaster management. Within this context, this study models transportation after the 6.8-magnitude earthquake that occurred on 24 January 2020 in Elâzığ, a region without tsunami risk, by using the four-step transportation modeling framework. The study is based on survey data collected from 2.739 earthquake survivors residing in 38 neighborhoods of the central district of Elazığ. In this study, face-to-face household surveys conducted between 9 August 2021 and 21 August 2021 were used. In line with the scope of this study, the aim of the survey was to collect data about the transportation preferences of individuals after the earthquake in Elazığ. In essence, the focus was on transportation demand. Accordingly, the formation and structure of the well-known household surveys were adopted to prepare the questions. A household survey in a standard transport demand analysis study usually enquires about daily travel on a typical day, while the survey in this study asked for the same in the first 24 h after the earthquake. The survey consisted of two parts. The first part contained 16 questions about personal, household, and building- and disaster-related attributes. The second part enquired about trip decisions within 24 h after the earthquake. The concept of the trip was described to the participants as“It is the movement from the place earthquake experienced to another place with a specific purpose, such as receiving news from relatives or friends, sheltering, reaching safe open areas, and shopping at gas stations, etc.”. All vehicle and walking movements longer than 300 m were considered to be a trip decision. In addition, questions were asked regarding the number of trips taken, the purposes of those trips, the departure and destination areas, the duration of the trips, and the type of transportation used.All the questions were open-ended. The majority of participants are women, adults, married individuals, and those without a driver's license. When household characteristics are examined, it is observed that the majority of households consist of those with a household size of 4 or more, households with at least one individual under the age of 18, households without any individuals aged 65 or older, households with at least one vehicle, and households without pets. When looking at the participants' housing, the number of participants living in apartment buildings, in buildings 20 years old or younger, who have resided in their current building for 1 to 10 years, and who are homeowners is higher. Seventy-two percent of participants do not know the locations of assembly areas, which are designated safe zones following an earthquake. On the other hand, individuals who had previously experienced a similar earthquake make up only 12% of the participants. When asked about their location at the time of the earthquake, 12% of the participants stated they were not at home. Since the earthquake occurred during the winter months and at night, 88% of the sample experienced the earthquake in their own homes. The analysis results indicate that 75% of the participants made at least one trip within the first 24 hours after the earthquake, and that these trips were temporally concentrated in periods close to the time of the earthquake. It was found that the number of trips made by participants ranged from 1 to 10, and a total of 3,477 trips were recorded. The distribution of trip frequencies indicates that 25% of the participants made no trips, 37% made one trip, 26% made two trips, 8% made three trips, and 4% made four or more trips. Of these trips, only 4% were made by public transport, while 37% were made by private car and 59% on foot. Prior to modeling the demand using four-step-transportation model, a binary lojit model was employed to investigate which attributes are associated with trip decisions after the earthquake. The study area was divided into regions based on perceived intensity. It was observed that household, building-and disaster-related attributes influence earthquake survivors' trip decisions. The initial location at the time of the earthquake was the most significant factor affecting trip decisions. It was also found that individuals who experienced the earthquake outside their homes in both datasets were more likely to make a trip. Additionally, the dataset with higher earthquake intensity had more significant variables affecting the trip decision In the model used in this thesis, the trip generation, trip distribution, and mode choice step of the classical four-step transportation model were applied, while desire lines were prepared for the traffic assignment step. In addition, several methodological modifications were introduced to adapt the conventional four-step transportation model to post-earthquake conditions. One of these modifications was the use of logit models in the trip generation step instead of traditional regression-based approaches. The primary motivation for this choice is that logit models incorporate unobservable and intangible factors through their random utility component. In extraordinary situations such as earthquakes, travel demand cannot be adequately explained solely as a linear function of socio-economic variables. Individuals' perceptions, constraints, and decision-making processes are considered essential elements in modeling such events and should not be neglected. Another methodological modification was applied in the trip distribution step, where calibration of the gravity model was not performed; instead, travel attraction probabilities were determined using a logit-based formulation. The mode choice model was constructed in the same manner as the classic four-step transportation model, using a logit model; however, it also includes earthquake-related variables among the explanatory variables. In the traffic assignment step, the use of desire links was considered more appropriate due to the significant uncertainties that may arise in the post-earthquake context. Across all steps of the model, the damaged building ratio variable was found to be statistically significant in the logit models. This finding indicates that the damaged building ratio is a key determinant of post-earthquake travel demand throughout the modeling process. In the generation model, the number of trips was defined as the dependent variable, while household size, the proportion of damaged buildings, and whether the individual is under 18 were defined as independent variables. In the attraction model, trip steps were selected as the dependent variable, and the proportion of damaged buildings and the presence of a hospital were selected as independent variables. In the distribution model, the catchment areas were defined as the dependent variable, while the proportion of damaged buildings and the ratio of the assembly area to the defined regional areas were defined as independent variables. In the mode choice model, vehicle types were defined as the dependent variable, while travel distance and the proportion of damaged buildings in the trip's origin area were defined as independent variables. In this study, each neighborhood was considered as a separate zone. The model results indicate that the highest number of motor vehicle trips within zones occurs in the Ataşehir, Cumhuriyet, and Sürsürü neighborhoods. It is therefore anticipated that transportation-related problems may arise in these zones in the post-earthquake period. In particular, given the high trip density in these neighborhoods, traffic congestion is likely to occur following the earthquake. With regard to motor vehicle trips outside the zones, demand is observed to be highest toward the Çaydaçıra neighborhood. Accordingly, traffic problems are expected to emerge along the routes providing access to this zone. The model analyses show that the Cumhuriyet neighborhood is the most densely traveled zone for intra-zonal walking trips. Similar to motorized trips, inter-zonal walking trips also show a high level of demand toward the Çaydaçıra neighborhood. This neighborhood is perceived as safe by earthquake-affected individuals and has therefore become a preferred zone for both motorized and walking trips. A large share of walking trips to this area originates from neighboring zones. Accordingly, the Çaydaçıra neighborhood has emerged as an important local attraction center in the post-earthquake period. Among the limitations of this study is that the developed model reflects travel behavior following an earthquake only during the winter season and at night, and only for earthquakes of a specific magnitude. This dissertation is one of the first studies in Türkiye to examine sudden post-earthquake travel demand within the framework of the four-step transportation model. It provides a behavior-oriented perspective for transportation planning in the disaster preparedness step. The findings are expected to support decision-makers in managing post-earthquake transportation more effectively and in developing appropriate policies and strategies accordingly.

Benzer Tezler

  1. Tez No

    846836

    Assessment of urbanization history of Addis Ababa city, Ethiopia

    Addıs Ababa cıty, Ethıopıa'nın kentleşme tarihinin değerlendirilmesi

    ABDURAHMAN HUSSEN YIMER

    Yüksek Lisans

    İngilizce

    İngilizce

    2023

    Şehircilik ve Bölge PlanlamaMersin Üniversitesi

    Şehir ve Bölge Planlama Ana Bilim Dalı

    DOÇ. DR. ALİ CENAP YOLOĞLU

  2. Tez No

    695018

    Development of decision support methods for restoration in earthquake-damaged medium voltage electrical distribution networks

    Depremden zarar gören orta gerilim elektrik dağıtım şebekelerinde restorasyon için karar destek yöntemlerinin geliştirilmesi

    MERVE BAYRAKTAR

    Yüksek Lisans

    İngilizce

    İngilizce

    2021

    Deprem MühendisliğiHacettepe Üniversitesi

    Yapı Ana Bilim Dalı

    DR. ÖĞR. ÜYESİ BURCU GÜLDÜR ERKAL

  3. Tez No

    733060

    Deprem sonrası sağlık kaynaklarının optimizasyon yaklaşımıyla planlanması: Kırıkkale kent örneği

    Planning health resources after earthquake with optimization approach: Kırıkkale city example

    SEMA ÇİFTÇİ

    Yüksek Lisans

    Türkçe

    Türkçe

    2021

    Endüstri ve Endüstri MühendisliğiKırıkkale Üniversitesi

    Endüstri Mühendisliği Ana Bilim Dalı

    PROF. DR. ÜMİT SAMİ SAKALLI

  4. Tez No

    948646

    Sıvılaşma kaynaklı yanal yayılma potansiyeli olan bölgelerde karayolları için kırılganlık eğrilerinin üretilmesi ve Küçükçekmece Avcılar D100 karayoluna uygulanması

    Development of fragility curves for highways in areas with liquefaction-induced lateral spreading potential and application to the D100 highway in Küçükçekmece Avcılar

    AHMET KAAN YILDIRIM

    Doktora

    Türkçe

    Türkçe

    2025

    Deprem Mühendisliğiİstanbul Üniversitesi-Cerrahpaşa

    İnşaat Mühendisliği Ana Bilim Dalı

    PROF. DR. İLKNUR BOZBEY

    PROF. DR. SADIK ÖZTOPRAK

  5. Tez No

    166663

    Deprem sonrasında uydu verileri ve yersel veri entegrasyonu ile arazi örtüsü/arazi kullanımı analizi

    Land cover/land use analysis after an earthquake through integration of satellite and ground data

    CİHANGİR AYDÖNER

    Doktora

    Türkçe

    Türkçe

    2005

    Jeodezi ve Fotogrametriİstanbul Teknik Üniversitesi

    Jeodezi ve Fotogrametri Mühendisliği Ana Bilim Dalı

    PROF.DR. DERYA MAKTAV

Geri Dön