Journal of Transportation Engineering
Year:2011 | Volume:2 | Issue:3 (7)|Papers Count:6

Uncertain demand is one of the most important sources of uncertainties in transportation networks. It is almost based on various sources such as difficulty in forecasting of socio economic variables and the model's deficiency. This demand is one of the main operative elements of transportation network design problems. The aim of this paper is to evaluate the effect of demand variability in discrete and continuous transportation network design problem. Thus, the uncertainty of demand has been assumed to have a maximum and a minimum rate, and the level of uncertainty in demand is assumed to vary between these extreme values as in a box. In this paper two Meta heuristic methods, Ant Colony Optimization (ACO) and Genetic Algorithm (GA), has been used to solve the aforementioned problem. It has been shown in this paper that relative to designing a network with maximum demand, the proposed method has found demands with 4 to 10 percent longer travel time. This could considerable effect on results of the network design problem.

Transportation of hazardous materials plays a significant role in economic activities in developing as well as developed countries because of the need of industrial plants to these materials. In many cases, production plants of hazardous materials are far from the consumers, where transportation of materials affects the safety of the people involved. In this paper, besides minimizing the cost of related transportation, reduction of accident probability as minimizing the risk of hazardous materials transportation, is also considered. The importance of required time for materials delivery is another main objective. The above issue is categorized in a vehicle routing problem with time windows (VRPTW). In this paper, a new multi-objective mathematical model is presented and solved by the use of a well-known evolutionary algorithm, NSGA-II. Related computational results are compared by results of Lingo software using the e-constraint method in small-sized test problems. A number of numerical investigations to evaluate the quality of solution sets versus time in large sizes are performed on various test instances. Results obtained by NSGA-II were compared based on two comparison metrics, namely spacing and quantity, in order to show the efficiency of the proposed NSGA-II. Finally, a real-case study was considered for shipping the Chlorine gas capsules from the Water & Wastewater Company to municipal water treatment facilities in Tehran.

Driving cycles are mostly developed in order to investigate emissions, fuel consumptions and/or legislations defined for a particular vehicle. On the other hand, automotive powertrains are designed or selected according to power demand analysis and simulations which are closely related to the vehicle driving pattern. When a car is utilized for a non-purposed application, its driving cycle is different so the design targets are not met. It results in lower efficiency and waste of resources, due to over or under engineering and lack of satisfaction. In this study, driving cycle of a typical passenger car is investigated in order to evaluate its design fitness. The driving cycle of the car is a combination of long modal trips which are assembled by large sets of transient micro-trips. It is constructed using actual data collected by a GPS and data logger. After filtering noises by LOWESS technique, statistical distribution of the dataset is verified so that Monte Carlo method can be implemented. Based on the acceleration results of the simulation, power demands of the vehicle and duty cycles can be derived and compared to the engine characteristics. The discrepancies indicate that the design of the vehicle powertrain does not match its application especially as a fix-route taxi.

Nowadays, railway has been identified as the safest and most proper transportation system. Failure of rails, as one of the most importance elements of railway systems, can cause accidents and affect its safety. In this study, 55 steel samples extracted from UIC60 rail have been tested by Charpy impact test equipment in different temperatures ranged between -40 and 60 C. Then, fracture toughness of rail in different temperatures has been calculated by using known experimental relation between fracture toughness and measured fracture energy. Finally, the rail critical crack length has been determined by replacing calculated fracture toughness in formula of stress intensity factor. In order to restrict rail fracture and increase service life of the rail, critical crack length, in addition to none destructive tests for maintenance process of rails.

Tehran Bus Transit System involves with two types of problems. The first type of problems is the performance-based problems, such as high fuel consumption, fast depreciation of the buses, … etc. Second type problems are spatial problems, such as unsuitable selection of bus routes and stops, the discontent of travel demand, … etc. In this paper, the efficiency of bus routes is evaluated by integrating Data Envelopment Analysis (DEA) model and Geographic Information System (GIS), from two viewpoints: performance-based and spatial problems. The parameters of number of bus stops, daily fuel consumption as input indexes and the number of passengers, the number of children and the aged people as well as the number of disabled people are selected as output indexes. The spatial data are prepared by GIS. Afterwards, the efficiency of decision making units (bus routes) is calculated using the DEA method and the more efficient routes are determined. In this model, the weights of the attributes are not required to be entered by the user, they are determined by DEA model, itself. After running the DEA model, spatial and performance-related efficiencies are calculated. The results show that, among the fourteen routes, five routes have high performance-related efficiency and three routes have high efficiency.

Line planning problem is a strategic and long-term decision in passenger railway planning. A line is a direct railway connection between two stations and is characterized by its origin and destination station, its frequency/day, the route between these two stations and the intermediate stops at passing railway stations. The line plan is the collection of all lines and, finding it, is a very complex problem. Researchers have used linear and non-linear mathematical programming for this problem. In this paper a new multi objective mixed integer linear mathematical model (MOMILP) for line planning problem is presented, in which two conflict objective functions, minimizing total operation costs and maximizing total number of direct passengers, are considered. Total operation costs consist of fixed cost, variable cost and direct passengers are the passengers that can travel from their origin station to their destination station without having to change trains. This model is solved by using goal-programming method. The companies of passenger trains use several kinds of rolling stocks that are different in speed and capacity, and the number of each kind of them is limited. Therefore, in the proposed model, the total number of existing rolling stock is limited by adding a new constraint, and more than one kind of rolling stock can be used for each line. Finally, a numerical example is presented to evaluate the performance of proposed model. The results show that by using goal programming, according to planning conditions and organizational priorities, both objective functions may be optimized simultaneously.