key: cord-0057627-afwf9n9o authors: Prathyusha, Challa; Singh, Sandeep; Shivananda, P. title: Strategies for Sustainable, Efficient, and Economic Integration of Public Transportation Systems date: 2021-03-18 journal: Urban Science and Engineering DOI: 10.1007/978-981-33-4114-2_13 sha: a7fc8c167d564b2bd4c234518b574641359a40cd doc_id: 57627 cord_uid: afwf9n9o In India, the city and intercity State Road Transport Undertakings (SRTUs) cater 6.8 million passengers/day over a length of 1.48 billion passenger-km/day during 2015–2016. The average occupancy ratio of 47 SRTUs has decreased by 1.56%, from 70.74 in 2014–2015 to 69.65 in 2015–2016. The traditionally followed operational plan for the public transportation system in urban areas results in a financial loss of 7% in SRTUs; from Rs. 10,587.98 crore in 2014–2015 to Rs. 11,349.78 crore in 2015–2016. In order to achieve the equalization of public service, there is a necessity for operational integration of the public transportation system to reduce the financial losses to public buses. The public transportation system productivity and efficiency could be maximized through the components of the Intelligent Transportation System (ITS) like Electronic Ticketing Machines (ETMs) and Global Positioning System (GPS) equipped with Vehicle Tracking Unit (VTU), which helps in the collection of data for a network. In this paper, the bus transit frequency is evaluated for one route with even headway by point check method. The bus service frequency software (Visual Basic-VB6) is developed to estimate user and operator costs for economic analysis. The results show that the waiting time cost of the passenger and the fleet size is minimized using deficit function as the solution for timetable synchronization and vehicle scheduling. In the context of Indian cities, the dependence of the urban trips on public transport is based on numerous parameters like the city size, geographical considerations, land use, and functional segregation of activities over the city. The Multiple Criteria Decision-Making (MCDM) method enhances the traveling behavior, and the provision of better transit service is challenging in the cities [1] . In India, the city and intercity State Road Transport Undertakings (SRTUs) cater 6.8 million passengers/day over a length of 1.48 billion passenger-km/day during 2015-2016. The average occupancy ratio of 47 SRTUs has decreased by 1.56%, from 70.74 in 2014-2015 to 69.65 in 2015-2016. The traditionally followed operational plan for the public transportation system in urban areas results in the financial loss of 7% in SRTUs, from Rs. 10,587.98 crore in 2014-2015 to Rs. 11, 349 .78 crore in 2015-2016 [2] . In urban areas, public transit remains to be a challenge due to an increase and variation in travel demand [3] . There is a necessity for public transport transit integration: organizational, physical, and operational, which can be achieved through push and pull measures [4] . If the demand for public transportation is less, then flexible service is desirable and conventional service method during high demand. The integration of these two methods develops a variable-type service [5] . In order to achieve the equalization of public service, there is a necessity for operational integration of the public transportation system to reduce the financial losses to public buses. The public transportation system productivity and efficiency could be maximized through the components of the Intelligent Transportation System (ITS) like Electronic Ticketing Machines (ETMs) and Global Positioning System (GPS) equipped with Vehicle Tracking Unit (VTU) which helps in the collection of data for a network [6] . Public transportation performance is checked through various service level benchmarks: headway, service coverage area, transit-accessible area, average waiting time, high-frequency accessible area, boarding, % transit ridership, organized Public Transportation (PT), peak hour, and availability of PT. The modified indicator was operating ratio, which is defined as the ratio between cost/km and earning/km [7] . According to [8] , the SRTU performance indicators in the Indian context were (1) Capacity-fleet size and vehicle seat capacity; (2) Serviceability-revenue per km, passengers per km, load factor, passengers carried, and number of routes; (3) Safetynumber of accidents, fatalities, and injuries; (4) Productivity-operating cost, cost per km, maintenance cost, traffic revenue, non-traffic revenue, operating ratio, and cost recovery; (5) Reliability-trip scheduled, regularity, actual trips operated, departure and arrival timings, number of breakdowns, and rate of breakdown; and (6) Comfortaverage age fleet were considered in the frequency and time table scheduling of the bus. The various factors influencing the efficiency of a bus operating system are reliability of service, frequency, route coverage, excessive transfers, fare and low profit [9] . Reliability of travel time is one of the concerned aspects for both user and operator and the bus arrival time will be influenced by various factors such as boarding and alighting of passengers, weather, and congestion [10] . However, to predict the travel time, weekly, daily, and hourly patterns need to be analyzed. The different arrival time prediction models are time series, regression, Kalman filtering, artificial neural networks, and super vector machines [11] . The time series model is used for the travel time of the bus at the stops. The bus service reliability is evaluated to improve the operational efficiency in two stages: strategic and tactical planning [12] . The 12 cities have been surveyed to know the satisfaction and dissatisfaction levels of the user from the bus service (primary mode-53.6%), and overcrowding is the most critical aspect of dissatisfaction; this can be improved through higher frequency [13] . The strategic-network design and the tactical-frequency setting and timetabling are given more importance in public transportation planning. The objective function, constraints, and decision variables of user and operator are used to optimize public transportation [14] . In Indian context, the public transport system productivity and efficiency could be maximized through the basic operational planning process. In this paper, the bus transit frequency is evaluated for one route with even headway by point check method. The bus service frequency software (Visual Basic-VB6) is used to estimate user and operator costs for economic analysis. The main objectives are as follows: • To develop a model for frequency setting for a route. • To develop a model for scheduling of bus for a particular route. The four basic operational planning processes include network route design, setting timetables, scheduling vehicles to trip, and assignment of drivers [15] . A costeffective frequency setting is one of the most complicated strategic planning for public transportation due to variations in travel patterns along the routes. The major service attributes route, operation, vehicle, user-level, and cost are considered with transit operators. The optimal frequency for different time periods of a single route for a single time period is determined considering user and operator attributes and parameters in "Bus Service Frequency Setting Software," which is mathematical programming in Visual Basic (VB6). The service efficiency constraints in the transit assignment such as cost per passenger, passenger per vehicle-kilometer, cost recovery ratio, number of trips per vehicle, number of passengers per trip per vehicle, operation cost per trip per passenger per vehicle with the objective to minimize the fleet size, and maximize the patronage. The regional bus scheduling problem [RBSP] is a network problem [NP] , which is solved in this paper using a bi-level programming model [16] and which helps in integrating the frequency and timetable of a single route [14] . The user and operator cost attributes influence the bus operating system. The overall cost is minimized considering two objective functions, minimum user and operator cost for a given route. The cost function is estimated using the handbook on feasible service delivery ranges for bus transit in the Indian Context part 2-operator perspective, chapter-4 for a given service period [17] as follows: Operator Cost Frequency is defined as the minimum number of vehicles requires for a given period [18] . In this work, the frequency is estimated for the baseline scenario using an hourly load point. (6) where α = Crowding level, h = Headway. The set of all trips arrives and departs within a scheduling horizon and does not consider Dead Heading (DH); the minimum number of vehicles required to serve all the trips is equal to the sum of all deficits. This is known as the fleet size theorem. Deficit function (DF) is defined as the necessity of a minimum number of vehicles required to complete N trips in the scheduling. In the scheduling, the change of the departure timings reduces the fleet size and waiting cost of the passenger. The failure of this method results in unbalanced loads, overcrowding, empty, and bus bunching problems, so DF is studied carefully while assigning vehicles to the trips. If the precedence relation R is satisfied, then vehicle is serviced sequentially as follows: A deficit function is also known as step function changes by −1 and + 1 for every arrival and departure times of each trip during the service period. Maximum{DF(j, h)|h = 1, 2, 3 . . . .H| (9) N Chain of the trip. V z Number of vehicles of size Z. Proof Given i = 1, 2, 3, 4 … V, each trip is assigned to i vehicles for Up and Down by the vehicle i, the required fleet size can be reduced where the flow C O can be saved by linking the trips together. Therefore, the minimum fleet size is required to perform all the trips. The historic data of passengers and capacity of the bus is considered to predict the passengers boarding and alighting. The predicted number of boarding passengers: After the departure of the bus (i), remain passengers from stop (j + 1). The number of passenger's in-vehicle: N i j+1 In-vehicle passengers for bus (i) after departure from stop (j + 1). N i j In-vehicle passengers for the bus (i) after departure from the stop (j). b i j+1 Boarding passengers for bus (i) at stop (j + 1). a i j+1 Alighting passengers for bus (i) at stop (j + 1). Check for bus capacity: C i j+1 saturation of capacity for bus (i) after departure from stop (j + 1). If C i j+1 ≤ 0 it presented that the number of passengers in the bus i reached the capacity. After departure from stop (j + 1), in-vehicle passengers for bus (i). Maximum number of passengers can be in the bus i. If the vehicle capacity of the bus is exceeded, then the remaining passengers are supposed to wait for the next bus. The remaining number of passengers can be calculated as where After the departure of the bus (i) from the stop (j + 1) remaining passengers. Pb i j+1 Passengers boarding for bus (i + 1) at stop ( j + 1). b i j+1 Passengers boarding for bus (i) at stop (j + 1). After departure from the stop (j + 1), a saturation of capacity for the bus (i). The number of alighting passengers can be estimated by passenger's in-vehicle: where pa i+1 j+1 predicting alighting passengers for bus (i + 1) at stop (j + 1). In-vehicle passengers for bus (i) after departure from stop (j + 1). σ j+1 % passengers alights at stop (j + 1). Bengaluru Metropolitan Transportation Corporation (BMTC) is one of the most dominant modes with a fleet size of about 6165 buses, including feeder services to metro covering 53,984 km catering 45 lakh commuters per day along 2194 routes with 43 depots during 2016-2017 [19] . The fleet utilization of BMTC buses is decreased by 7.1% from 2013-2014 (91.2%) to 2018-2019 (84.1%). The revenue of BMTC for the non-air conditioned bus is Rs. 41.48/km, whereas the operating cost is Rs. 57.88/km [20] . Hence, there is a necessity for operational integration to reduce the loss of BMTC buses. The selected bus route-290E goes along the route Yelhanka to Shivajinagar, located in the northern part of Bangalore city. It has a route length of around 20.1 km with 13 major bus stops. The input details for morning peak hour 08:00:00 a.m. to 08:59:00 a.m. in the software are given in the tables as follows (Tables 1, 2, 3 and 4): The model and software developed balances the bus service operation for different service periods considering upper and lower constraints. Figures 1 and 2 1 0 5 10 15 19 20 20 22 22 22 23 23 24 2 5 0 5 10 15 19 20 20 22 22 22 23 23 3 10 5 0 5 10 15 19 20 20 22 22 22 23 4 15 10 5 0 5 10 15 19 20 20 22 22 22 5 19 15 10 5 0 5 10 15 19 20 20 22 22 6 20 19 15 10 5 0 5 10 15 19 20 20 22 7 20 20 19 15 10 5 0 5 10 15 19 20 20 8 22 20 20 19 15 10 5 0 5 10 15 19 20 9 22 22 20 20 19 15 10 5 0 5 10 15 19 10 22 22 20 20 20 19 15 10 5 0 5 10 15 11 23 23 22 22 22 20 20 19 10 5 0 5 10 12 23 23 22 22 22 20 20 19 15 10 5 0 5 13 24 23 23 22 22 20 20 20 20 15 10 5 0 The study results point out the frequency of a single route for a single time period considering user and operator costs. The simulation-based frequency through mathematical programming considering operator and user service constraints caters to the sustainable, efficient, and economical public transportation system, which saves resources and provides an environment-friendly transportation system in urban areas. The waiting time cost of the passenger and the fleet size is minimized using the Deficit function as the solution for timetable synchronization and vehicle scheduling. The objective function is the same for pre-and post-COVID-19. However, the desired occupancy constraint values vary under social distancing, frequency, and timetable synchronization of a particular route that can be evaluated using this model based on demand variation. Thus, the routes are rationalized based on the variation in boarding and alighting of the passengers between the origin and destination. Planning of an integrated urban transportation system based on macro-simulation and MCDM/A methods Review of the performance of state road transport undertakings for Uncertainty in travel demand forecasting: Literature review and research agenda Public transport integration: The case study of Thessaloniki Conventional, flexible, and variable-type bus services Estimation of bus arrival times using APC data Service level benchmark in urban transport for Indian cities volume-I-benchmarking manual Indicators to measure performance efficiency of bus systems final report Grant Agreement Ref Modeling bus travel time reliability with supply and demand data from automatic vehicle location and smart card systems A prediction model of bus arrival time at stops with multi-routes Simultaneous bus transit route network and frequency setting search algorithm User satisfaction with city bus public transport in India Public-transport automated timetables using even headway and even passenger load concepts An improved model for headway-based bus service unreliability prevention with vehicle load capacity constraint at bus stops. Discrete Dynamics in Nature and Society Optimization of bus route planning in urban commuter networks Handbook on feasible service delivery level ranges for bus Bus frequency determination The Economic Times