Please use this identifier to cite or link to this item:
Title: การวิเคราะห์ความถี่และตารางเวลาการให้บริการของระบบขนส่งสาธารณะภายในมหาวิทยาลัยเชียงใหม่
Other Titles: Analysis of transit frequency and scheduling in Chiang Mai university
Authors: วณัฐตรา สว่างใจ
Authors: ปรีดา พิชยาพันธ์
วณัฐตรา สว่างใจ
Issue Date: May-2021
Publisher: เชียงใหม่ : บัณฑิตวิทยาลัย มหาวิทยาลัยเชียงใหม่
Abstract: The objective of this research was to analyze the current service frequency, to determine the preliminary service frequency. and to scheduling services in the Chiang Mai University public transport system (CMUT) where contains five routes within the Suan Sak side and have a total of 43 stations. The number of passengers in this research obtained from the tracking and passenger counting system on the CMUT electric shuttle. This research was divided into 3 steps: In the process of analyzing the number of passengers and related information of the CMUT electric shuttle service. From Monday to Friday, the time periods with the highest passenger number are class change, lunch breaks, and after school periods, and lowest passenger number are late-night periods.From Saturday to Sunday, the peak hour is in the evening, and the least passenger time is in the morning and late night. The route with highest passengers is Line 2. Meanwhile, the route with the least number of passengers is Line Express. The station where presents least number of passengers is the passenger pavilion where located before the male dormitory 6. The station where presents highest service is CMUT intersection stations 1-4 located opposite to the female dormitory 3. The density of the load profile is from 0.14 to 0.65. Therefore, all four methods are defined to calculate the initial service frequency. And, from most of the daily passenger-period graphs, presented trendlines which have a slope greater than -1.00. In the forecasting process, there are three patterns of the data, first pattern is Daily data, second pattern is Weekdays and Weekend, and third pattern is according to the class schedule. Data in each pattern has a similar number of passengers, but there may be a period that the value is slightly different from most of the data. The calculation procedure has processed by 13 methods and 3 forecast error measurements: MSE, MAE, and MAPE. From the analysis, it was found that Appropriate methods andpatternfor each time are different. But to simplify its implementation, the classical multiplicative decomposition method by calculating the trendline with the least squares method first (CL) is used during peak periods and trend-adjust exponential smoothing (HES) method (weekdays) and linear regression analysis (LR) method (Weekend) is usedin off-peak periodand use third pattern,arranging the service days according to the study schedule. This pattern is frequency used method in this study In the process of determining the preliminary service frequencies in all 7 cases, it was found that the most of the current service frequencies were greater than the number of passengers, leading to empty seats in vehicles. But in some periods, the current service frequencies were less than the number of passengers, resulting in insufficient vehicles and in some periods the current service frequency is suitable for the number of passengers by considering the percentage of the difference between service frequency obtained and the current service frequency and the percentage of service frequency achieved are equal to the current service frequency in case 1, the policy frequency equals to the current service frequency. The case that is suitable for use in service scheduling is case 5, the policy frequency is 3 cars/hour, which is determined from the difference of the percentage difference of service frequencies obtained with current service frequencies, and policy frequencies, and a simple weighted integration (SAW) approach is used to aid decision-making. In each process of timetabling each route, the frequency determination method will be selected based on the number of departures and minimum vehicle fleet size anduse the even-headway method. The timetable recommended in this research chose method 2 during off-peak hours. Because it has the highest number of departures but can accommodate the number of passengers appropriately, and the minimum vehicle fleet size of method 2 is equal to method 1 and 4. Method 4 is used for peak hours. Because it is the frequency method that can determine the frequency during the hourly max load and has the number of departures more than method 3. The recommended service frequency during the peak period was 4.3 - 16.5 cars/hour (h= 3.5 – 15 minutes) for Monday-Friday and 3.4 – 6.5 cars/hour (h = 10 – 18.5 minutes) for Saturday-Sunday. and during off-peak period was 3.1 - 8 cars/hour (h = 9 - 19.5 minutes) for Monday -Friday and 3 - 3.6 cars/hour (h=17 - 20 minutes) for Saturday –Sunday
Appears in Collections:ENG: Theses

Files in This Item:
File Description SizeFormat 
590631094 วณัฐตรา สว่างใจ.pdf27.78 MBAdobe PDFView/Open    Request a copy

Items in CMUIR are protected by copyright, with all rights reserved, unless otherwise indicated.