F O Montgomery

A STUDY OF HYBRID NEURAL NETWORK APPROACHES AND THE EFFECTS OF MISSING DATA ON TRAFFIC FORECASTING

In this paper we present an application of hybrid neural network approaches and an assessment of the effects of missing data on motorway traffic flow forecasting. Two hybrid approaches are developed using a Self-Organising Map (SOM) to initially classify traffic into different states. The first hybrid approach includes four Auto-Regressive Integrated Moving Average (ARIMA) models, whilst the second uses two Multi-Layer Perception (MLP) models. It was found that the SOM/ARIMA hybrid approach out-performs all individual ARIMA models, whilst the SOM/MLP hybrid approach achieves superior forecasting performance to all models used in this study, including three naïve models. The effects of different proportions of missing data on Neural Network (NN) performance when forecasting traffic flow are assessed and several initial substitution options to replace missing data are discussed. Over-all, it is shown that ARIMA models are more sensitive to the percentage of missing data than neural networks in this context.

Car-Following Model for Motorway Traffic

This paper presents a new car-following model that aims to capture some of the key motorway flow characteristics, namely, traffic breakdown, hysteresis, and shock wave propagation, as well as close-following behavior. The model proposes three different driving states: nonalert, alert, and close following. Under the different driving states, drivers apply different reaction times and accelerations. This paper presents the formulation and algorithmic implementation of the model. The theoretical analysis of the macroscopic flow-density relationships of the model is discussed. Simulation experiments were conducted, and the results are examined at both the macroscopic level (speed breakdown and traffic hysteresis) and the microscopic level (gap distribution and shock wave propagation). The results show that the model is able to capture realistically the speed drop, traffic hysteresis, and shock wave propagation as well as close-following behavior. Further studies of the sensitivities of key model parameters su...

COMPARING THE SPEED-FLOW RELATIONSHIP FOR MOTORWAYS WITH NEW DATA FROM THE M6

This paper draws on data collected in March 1993 from the M6 motorway north of Birmingham, England, to discuss the speed-flow relationships used in cost benefit analysis. Piece-wise linear functions have been fitted to the M6 data, and are then compared to the relationships in the U.K. Cost Benefit Analysis Manual (COBA9: U.K. Department of Transport, 1981), and the U.S. Highway Capacity Manual (HCM: U.S. Transpn Res Bd, 1994). The results provide clear support for the value of 1200 v/h/l used in COBA9 as the flow at which speeds start to decrease more rapidly, but in other respects the results differ from values in both COBA9 and the HCM. In comparison with the observed data, COBA9 estimates: a lower speed at a flow of 1200 v/h/l; a steeper slope at higher flows; a higher capacity flow; and a lower speed at capacity. Equivalently, HCM estimates: a gentler slope at higher flows; a higher capacity flow; and a higher speed at capacity. Thus the high-flow data fall somewhere between the COBA9 and HCM estimates.

Modeling Bus Lane Priorities in a Motorcycle Environment Using SATURN

Bus lanes and median bus ways such as those in Curitiba, Brazil, and Bogotá, Colombia, are attracting a great deal of attention worldwide. For many Asian developing cities, there is an interest in how successfully bus lanes and bus rapid transit will function, given that traffic conditions in such cities typically are different from those found in Latin America. This paper addresses this issue for Asian cities whose traffic is dominated by motorcycles, concentrating particularly on Hanoi, Vietnam. With the use of a SATURN mesoscopic simulation model for Hanoi (with 322 junctions and 1,108 links), different bus lane–bus way designs are compared in regard to their effects on travel time savings. Results show that the level of motorcycle violations has an important impact on the success of bus lane schemes and that there is no significant speed improvement on bus lanes if enforcement is weak. Bus ways can achieve much higher bus speeds than can bus lanes, but general traffic speeds are likely to reduce them significantly if there is no mode switching from private modes to buses. In such circumstances the potential extra delay due to a poorly designed bus way is greater than that due to a poorly designed bus lane. The route-switching flexibility of motorcycles also has considerable effects on the performance of bus lanes and bus ways.

Integrated Traffic Control On Urban Arterials

This paper is based on the DRIVE project PRIMAVERA, the main aims of which are to develop, test and produce recommendations for the application of integrated traffic control and management measures incorporating: queue control, public transport priority, and environmental protection. Strategies will be be tested in the two case study cities of Leeds and Torino.

THE SIMULATION OF VARIABLE TRAFFIC SIGNAL CONTROL IN THE BANGKOK NETWORK

Due to the rapidly rising car ownership in Bangkok, the capital of Thailand, traffic congestion is increasing rapidly. TRAF‐NETSIM was selected for the detailed analysis of Bangkok conditions. However, most traffic signals in Bangkok are controlled by on‐site traffic police operating the controllers in manual mode. Hence the cycle time, length of stages and the sequence of stages are dependent on the minute by minute decisions of the traffic police. One of the constraints of TRAF‐NETSIM (and most other microsimulation models) is that the cycle time can have only one value for the whole network, and the sequence of stages must be in the same order throughout the simulation period. Therefore, the varying cycle times operated by the police had to be represented as a common value, and the sequence of stages had to be fixed. It was then necessary for simulation purposes to adjust the green times to maintain the observed capacities. The principles in Webster and Cobbe (Traffic signals, Technical Paper No. 56, R...