Content uploaded by Guo Rongge
Author content
All content in this area was uploaded by Guo Rongge on Jun 22, 2021
Content may be subject to copyright.
A preview of the PDF is not available
Time-Dependent Urban Customized Bus Routing With Path Flexibility
Abstract and Figures
Urban customized bus companies are increasingly motivated by design efforts that entail more efficient route scenarios to incorporate adaptation to temporal and spatial hetero-geneity in travel demand. However, such motivations are usually hindered by ubiquitous arrival unpunctuality resulting from traffic congestion. To resolve this problem, we suggest a time-dependent bus route planning methodology that explicitly considers path flexibility between nodes to be visited. First, we establish a mixed-integer programming model to formulate the problem, where decision-making considerations in bus route planning, path choice between nodes, and passenger assignment are concurrently integrated. Then, we develop a hybrid metaheuristic (combining tabu search and variable neighborhood search) to solve the model, in which satisfactory performance is observed from the numerical test in a small-sized example. Finally, the problem and methodology are addressed in a city-scale instance, where the effects of time-window features and traffic congestion, as well as the benefits from path flexibility inclusion in terms of cost, travel time, and distance are investigated.
Figures - uploaded by Guo Rongge
Author content
All figure content in this area was uploaded by Guo Rongge
Content may be subject to copyright.
... The heterogeneous fleet solution minimized 30% of the total cost and enabled the utilization of all the 105 buses' capacity. 106 107 Machine-scheduling [29] School decomposition algorithm [31] Mixed-load Improvement Algorithm [45] Augmented Lagrangian Relaxation [14] Metaheuristic Evolutionary Genetic Algorithm (GA) [5,19,35,32,43,38,26,44] Ant Colony Optimization (ACO) [11,44,52] Estimation Distribution Algorithm (EDA) [19] Intelligent Water Drops Algorithm (IWD) [3] Local Search Tabu Search (TS) [1,15,25] Hybrid iterated local search (ILS) [23,39,51] (Iterated) Local Search (LS) [22,41] Simulated Annealing (SA) [13,47] Constructive Greedy Randomized Adaptative Search Procedure (GRASP) [50] Hybrid GA + TS [46] GAA: GA + ACO [49] GA + Exact +TS and SA+Exact+TS [28] GA+ILS [30] ILS + SP [40] TS+VNS [54] AAC+VNLS [24] GRASP + VND [10] ACO+ ILK [36] 109 Stochastic approaches are also used with exact methods. In [55], numerical experimentation was performed 110 with instances with zero transition time, non-zero transition time, and multiple scenarios. ...
... Finally, a hybrid metaheuristic method is developed in [54], which combines TS and the VNS. The study 240 ...
... The cost is a function of delays and the disutility of travel times [48]. In [54], the approach considers 272 path flexibility between nodes to be included in the route. It combines TS and VNS to solve the model 273 and evaluate the effects of congestion on cost, travel time, and distance. ...
The School Bus Routing Problem (SBRP) is a classic optimization problem with a massive potential for real applications that have a high impact on society. Research interest in this problem is constantly rising. Previous review papers with a time-space between them of 10 years have helped understand the different features studied by the research community about this problem. In this systematic review, we consider two new categories not discussed before: the incorporation of a mixed load composed of multiple schools, along with the inclusion of a smart element related to the availability of user information and communication in real-time to join the smart mobility trend. In addition, we discuss the lack of real applications in the SBRP in university contexts with a focus on the multi-load problems.
... Through the establishment of fixed operational passenger lines and timetables, a greater variety of choices and convenience has been provided for existing intercity carpooling passengers. Although travel safety, fare prices, and travel time windows are considered crucial for passenger choices regarding road transport [15][16][17][18][19], numerous studies indicate that an effective service plan (including stations, lines, and timetables) is the core of the entire operational system. Its quality directly impacts passengers' individual experiences and determines the feasibility of entering the market [4,6,20]. ...
... Let the new solution as L 0 and the current solution as L, and the initial temperature of metropolis as IP as well as the temperature variation rate as TK. Calculate the difference between the total profit values of the new solution and the current solution DTP, as shown in the formula (18). Temperature control is an important component of the metropolis rule, which adjusts the probability of accepting a worse solution to gradually decrease, thus allowing the algorithm to gradually converge to the global optimal solution. ...
Traditional intercity passenger transportation is inefficient, inflexible, and financially unrewarding, failing to meet the demands of intercity travel. To address these issues, this study utilizes historical carpooling order data, extracts travel patterns, and devises a comprehensive plan for intercity customized passenger transport services. Firstly, a formalized description of issues related to long-distance cross-city travel, multiple lines, and scheduling around highway entry and exit points is addressed. Secondly, a single-objective integer linear programming model is constructed, aimed at maximizing the total profit for the operating company. Finally, from a spatial-temporal network perspective, a refined balanced iterative reducing and clustering using hierarchies (BIRCH) algorithm is designed for alternative station selection. To achieve a rapid and effective solution to the model, the approach is combined with a variable neighborhood search algorithm. Experimental results on historical carpooling data from Anxi County and Xiamen City demonstrate that the proposed algorithm, compared to the combination of BIRCH and genetic algorithms, exhibits shorter computation time, higher quality, and stability. Additionally, various parameter analysis and sensitivity experiments demonstrate the effectiveness of parameters used.
... One port is randomly set on the edge of the sea area. The proposed LKH-TPM1 and LKH-TPM2 are compared with four baseline algorithms: the ant colony optimization (ACO) method in [18], the simulated annealing (SA) algorithm in [24], the taboo search (TS) algorithm in [25] and greedy local search (GLS) algorithm (the SA no longer accept with certain probability solutions that are inferior to the current optimum). The parameter settings of these algorithms are introduced as follows. ...
Due to the high mobility of Unmanned Aerial Vehicle (UAV), it can be an effective method for pollution detection of vessels on the sea. How to optimize the flight path of the UAV so that the visited energy consumption is minimized is a problem that remains to be solved. In this paper, the Lin-Kernighan-Helsgaun-based trajectory planning method (LKH-TPM) is used to solve the UAV scheduling problem to minimize the UAV visit path length and compare it with the ant colony (ACO) algorithm, simulated annealing (SA) algorithm and tabu search (TS) algorithm. The experiments are carried out under different ship numbers, different sea areas, and different base station numbers, and it is verified that LKH-TPM is a more effective solution for the problem under study.
Artificial intelligence (AI) is an innovative concept that can provide potentials to overcome the challenges of operation in public transport (PT) systems. The AI applications in the PT operation offer the opportunities to alleviate traffic congestion and enhance the accessibility, mobility, and reliability of services with a more efficient and effective PT system. One of the key areas that AI is beneficial is in optimization of network route design. Examples of AI approaches that are finding their way under fuel and electric vehicle conditions include genetic algorithm (GA), simulated annealing (SA), etc. The more promising application of AI techniques requires a well-understood knowledge of multiple data sets and features of different PT services, including conventional buses and demand-responsive transit systems, especially when dealing with travel demands fluctuating in time and space. Moreover, the emerging development in connected and autonomous vehicles (CAVs) is leading a rapid improvement in flexibility, punctuality, vehicle safety, and transit priority. The purpose of this chapter is to review AI approaches applied on public transport operation, especially on routing. The overview concludes by addressing the issues and challenges of AI applications in PT operation.
As a new public transport service mode, customized bus (CB) can meet the diversified and personalized travel needs of residents. However, it is difficult for bus operators to make precise customizations. In this paper, considering the characteristic of variable CB stations and passenger demand, a mixed integer programming model is developed for the CB bus route design, and a two-stage heuristic with CW saving algorithm and variable neighborhood search is proposed. The proposed methodology is validated using a real-world transit network in Beijing. The results show that the occupancy rates of the most CB bus lines designed in this paper can reach 100%, and the average occupancy rate can reach 91.67%.
The demand-responsive customized bus has been operated in real life, which is a crucial way to improve the service quality and efficiency of the urban public transportation system. Reasonable station and line planning can enhance customized bus competitiveness in residents’ travel mode. Most previous studies on optimizing customized bus lines rely on historical passenger volume and travel time to generate static schemes, but the actual operation process of customized bus is often in uncertain circumstances, such as road congestion. The static strategy will occur deviations in this situation. This study proposes a novel planning method to address the above issue. First, a spatiotemporal clustering algorithm is proposed to generate joint stations based on the passenger travel demand. Second, the method models the multiline customized bus optimization problem as a Markov decision process and uses a multiagent deep reinforcement learning algorithm to ensure effective training and response to incomplete information scenarios. Finally, the rationality of the proposed planning method is verified in a case study of customized bus area in Chongqing, China. Compared with the latest heuristic optimization algorithm, our method can effectively reduce the operating and passenger costs in complex environments.
This paper presents a novel framework for customized modular bus systems that
leverages travel demand prediction and modular autonomous vehicles to optimize
services proactively. The proposed framework addresses two prediction scenarios
with different forward-looking operations: optimistic operation and pessimistic operation.
A mixed integer programming model in a space-time-state network is developed
for the optimistic operation to determine module routes, schedules, formations
and passenger-to-module assignments. For the pessimistic case, a two-stage
optimization procedure is introduced. The first stage involves two formulations (i.e.,
deterministic and robust) to generate cost-saving plans, and the second stage adapts
plans with control strategies periodically. A Lagrangian heuristic approach is proposed
to solve formulations efficiently. The performance of the proposed framework
is evaluated using smartcard data from Beijing and two state-of-the-art machine
learning algorithms. Results indicate that the proposed framework outperforms the
real-time approach in operating costs and highlights the role of module capacity and
time dependency.
Detours are inevitable in on-demand customized bus (CB) systems. Previous studies alleviate the impact of detours by predefining high spatial-temporal similarity of travel orders for CB. However, this assumption is clearly inconsistent with orders’ distribution at the urban level and leads to low bus occupancy rate in practical use. In this paper, we propose a novel service policy to achieve cost-effective customized bus, which consists of dynamically deployed bus stops and a spatial-temporal heterogeneous-order service. Then, to address the challenge of computational complexity, we provide an order-oriented graphic model named order correlation network (OCN) to formulate the CB design problem. By introducing OCN, we propose a near-optimal computationally efficient solution to the problem, which is scalable and suitable for real-time implementation. Finally, comparative experiments based on the real-world taxi trajectory dataset in San Francisco are implemented to verify the performance of our proposed CB in terms of service coverage and travel efficiency.
Shared transport is an emerging transportation system where travelers share a vehicle either simultaneously as a group (e.g. ride-sharing, ride-sourcing, or customized bus) and share the cost of the journey in the process, or over time (e.g. car-sharing or bike-sharing) as personal rental. We review shared transport research articles published in major operations research, management science and transportation journals from 2010 to 2021. In reviewing these studies, our objective is to assess and present the operations management research of shared transport from a macro level such that new researchers will be attracted and established researchers will be motivated to contribute to this emerging field. We divide the operations management research of shared transport into eight attributes (shared transport mode, decision level, characteristics of demand, decision function, decision objective, modeling approach, solution method, and type of data). Cross tabulations are done among these eight attributes to gain more profound insights. Furthermore, we divide the decision function attribute into 36 subcategories. Maturity and recent attention are calculated as indicators for evaluating the research potential in the decision function attribute. After analyzing the research status, we look forward to the prospect and provide relevant recommendations for future research.
Considering that, in reality, passengers usually hold preferred time windows when waiting for a bus at a station, this study develops a mixed integer programming model for the customized bus routing problem (CBRP) with full spatial–temporal constraints based on one of our previous studies. Specifically, bus routing and passenger assignment are simultaneously optimized with better vehicle capacity utilization and more realistic considerations of partial service, characteristics of customized bus service, and a range of operational constraints. To solve the formulated model, an exact algorithm (i.e. the branch-and-cut algorithm) and two heuristics (i.e. the genetic and tabu search algorithms) are numerically compared through an illustrative example, and subsequently, a case study in Beijing is conducted to assess the proposed approach. A comparison with the practical customized bus system shows that the proposed approach realizes effective vehicle usage on several routes.
In recent years, the customized bus (CB) has been introduced and popularized in China to improve the attraction and service
level of public transportation. A key point of the CB system, the route design problem, is always formulated as a vehicle routing problem
with pickup and delivery (VRPPD). However, VRPPD cannot sufficiently describe the in-vehicle passengers of multiple vehicles involved.
In this paper, a mixed integer programming model is developed to formulate a multivehicle routing problem, with suggestions for bus stop
locations and routes. Meanwhile, the model can determine passenger-to-vehicle assignment based on a series of constraints, like operation
standard and number of stations. In solving the problem, a numerical example is used to compare a genetic algorithm (GA) and branch-andcut
algorithm. The comparison results illustrate that GA is more efficient with lower complexity. Finally, in order to apply and evaluate the
proposed model, a real-world case study using smartcard data is conducted to compare the approach with the current CB route design
method in Beijing.
Custom bus routes need to be optimized to meet the needs of a customized bus for personalized trips of different passengers. This paper introduced a customized bus routing problem in which trips for each depot are given, and each bus stop has a fixed time window within which trips should be completed. Treating a trip as a virtual stop was the first consideration in solving the school bus routing problem (SBRP). Then, the mixed load custom bus routing model was established with a time window that satisfies its requirement and the result were solved by Cplex software. Finally, a simple network diagram with three depots, four pickup stops, and five delivery stops was structured to verify the correctness of the model, and based on the actual example, the result is that all the buses ran 124.42 kilometers, the sum of kilometers was 10.35 kilometers less than before. The paths and departure times of the different busses that were provided by the model were evaluated to meet the needs of the given conditions, thus providing valuable information for actual work.
The development of bike sharing system (BSS) has changed travelers’ commuting and lifestyle in recent years. Whether BSSs are complementary or competitive to public transit remains controversial. This study uses a propensity score matching-based difference-in-difference (DID) method to evaluate the impact of free-floating BSS on bus ridership in Chengdu, China. The transaction data of bus service and BSS and the neighboring points of interest are investigated. Results indicate that, (a) on the bus route level, each shared bike results in a 4.23 increment and a 0.56 reduction in daily bus ridership on weekdays and weekends, respectively; (b) on the bus stop level, the increment in shared bikes significantly negatively impacts bus ridership on weekends; (c) on the route level, regarding the time of day, each unit increment of shared bike significantly increases bus ridership on weekdays by 0.54, 0.34, and 0.15 during a.m. peak, p.m. peak, and off peak, respectively; and (d) on the bus stop level, the relationship between shared bikes and bus ridership is insignificant on weekdays. This study reveals that the demand pattern of commuters strongly impacts the relationship between shared bike and public transportation.
Bus holding is a widely used control method to regularize bus headways and reduce bus bunching. The method works in such a way by delaying buses at control points if their departure times or headways deviate from the planned ones. However, it may result in reduced bus commercial speeds and increased passenger onboard travel time. To avoid this problem, researchers have suggested that control points be spaced cautiously along the route such that only a few are needed. This study proposes a predictive headway-based bus holding strategy with dynamic control point importance ranking and selection based on the cooperative game theory. The framework considers not only individual control points' impact but also the collective group control effects. Specifically, the proposed framework consists of two components: a performance model and a cooperative game model. The performance model predicts headway performances of all running buses when different control point combinations are in effect. Dynamic bus running times and passenger demands are reflected in the model. Then, these headway performances are passed to the cooperative game model with control points being players and improvements in headway performances compared with that under no holding control being the utility function. The game is solved by Myerson value, a concept that extends Shapley value used for the normal cooperative game and considers the cooperation structure and potential worth of coalitions. We use Myerson value to rank the importance of control points on regularizing headways, as it measures the average marginal utility contribution of a control point to all possible coalitions that exclude that point. We prove that Myerson value lies in the-core of the game and thus satisfies allocation efficiency, individual and coalition rationality. The proposed framework is applied to target headway control and two-way-looking self-equalizing headway control. Simulation results show that the framework can significantly reduce passenger waiting time and bus headway variation.
A customized bus (CB) system is an emerging public transportation that aims to provide direct and efficient transit services for groups of commuters with similar travel demands. Existing CB systems aggregate similar travel demands and plan bus lines manually, which is inefficient and costly. In this paper, we propose a CB line planning framework called CB-Planner, which is applicable to multiple travel data sources. A mathematical programming formulation is proposed to simultaneously optimize bus stop locations, bus routes, timetables and passengers’ probabilities of choosing CB. We then developed a heuristic solution framework that includes a grid-density based clustering method for discovering potential travel demands efficiently, a bus stop deployment algorithm to minimize the number of stops and walking distance, and dynamic programming based routing and timetabling algorithms for maximizing estimated profit. We conduct an experiment on a small-scale network to verify the performance gap between the optimal solution and our proposed heuristic solution. A case study is then conducted on one-month taxi trajectory data in Nanjing, China. The study demonstrates that CB lines generated by our CB-Planner can achieve higher profit compared with baseline methods, and they also provide efficient transit services with short walk distances and small departure time adjustments. The moderate increase in travel time is paid off by the significant savings in travel fare.
Customized commuter bus (CCB) is a kind of innovative public transit service launched starting in 2013 in many cities all over the world. It is designed to meet the direct travel demands of commuters who have similar origin and destination locations during their long-distance commuting trips at peak hours. To identify the origin and destination distribution of potential CCB passengers, this paper proposes a pair wise density-based spatial clustering algorithm. With the proposed algorithm, the method to extract the potential CCB passengers from regular bus passengers based on the bus smart card data is introduced. Meanwhile, the discovered hot locations of potential CCB passengers can be regarded as the candidate locations of CCB stops and can be used to set candidate CCB lines. Finally, the demand survey data collected from the actual registered CCB passengers are applied to verify the accuracy and feasibility of the clustering results obtained by the proposed algorithm. The related findings could guide bus operators to design CCB lines and allocate vehicle capacities on different lines.
Emerging transportation network services, such as customized buses, hold the promise of expanding overall traveler accessibility in congested metropolitan areas. A number of internet-based customized bus services have been planned and deployed for major origin-destination (OD) pairs to/from inner cities with limited physical road infrastructure. In this research, we aim to develop a joint optimization model for addressing a number of practical challenges for providing flexible public transportation services. First, how to maintain minimum loading rate requirements and increase the number of customers per bus for the bus operators to reach long-term profitability. Second, how to optimize detailed bus routing and timetabling plans to satisfy a wide range of specific user constraints, such as passengers’ pickup and delivery locations with preferred time windows, through flexible decision for matching passengers to bus routes. From a space-time network modeling perspective, this paper develops a multi-commodity network flow-based optimization model to formulate a customized bus service network design problem so as to optimize the utilization of the vehicle capacity while satisfying individual demand requests defined through space-time windows. We further develop a solution algorithm based on the Lagrangian decomposition for the primal problem and a space-time prism based method to reduce the solution search space. Case studies using both the illustrative and real-world large-scale transportation networks are conducted to demonstrate the effectiveness of the proposed algorithm and its sensitivity under different practical operating conditions.
We introduce the time-dependent capacitated profitable tour problem with time windows and precedence constraints. This problem concerns determining a tour and its departure time at the depot that maximizes the collected profit minus the total travel cost (measured by total travel time). To deal with road congestion, travel times are considered to be time-dependent. We develop a tailored labeling algorithm to find the optimal tour. Furthermore, we introduce dominance criteria to discard unpromising labels. Our computational results demonstrate that the algorithm is capable of solving instances with up to 150 locations (75 pickup and delivery requests) to optimality. Additionally, we present a restricted dynamic programming heuristic to improve the computation time. This heuristic does not guarantee optimality, but is able to find the optimal solution for 32 instances out of the 34 instances.