A mathematical model of traffic segregation on multilane road

Received Dec 21 th , 2018 We study a process of segregation of traffic flow on a multilane highway. Our study is based on the stochastic mathematical model developed by Buslaev A.P. et al. in 2006, 2008 years. This model is based on the deterministicstochastic approach that was developed by Buslaev in 2005. In the model, particles, moving on a lattice, correspond to vehicles. We consider the problem of optimization of segregation characteristics. We consider an example of segregation process on five-lane segment of a road. The length of the segment equals 4000 meters. The calibration of the model parameters is based on empirical data obtained in measurements on roads. Keyword:


Introduction
Currently, the network of main streets and highways, despite the total length of less than 5% of the total length of the road network, provides up to half of the total mileage of cars.For example, the Federal road system of the United States, with a length of 1.6% serves 23.8% of the total mileage, and within the influence of large and largest cities in the United States on high-speed roads account for 30 to 50% of the mileage, [1] (Mikhailov A., 2004).In Russia, the main road network accounts for 3.2% of the total length of roads, but performs more than 30% of freight and passenger traffic (Babkov V., 1993).The huge role of the network of main roads lies in ensuring the quality of transport services to the population.To highways impose high technical requirements, they are designed to ensure the movement of vehicles in continuous mode with high speeds.Geometrical parameters of the plan, longitudinal and cross-section of segments of highways allow, along with ensuring high speeds and indicators of traffic safety, to provide and the maximum possible indicators of their capacity (Silyanov V., 2008).In addition to the motorway segments, the most important elements of motorways, which often determine their capacity, are the merge and diverge areas of freeway interchange -interchange influence areas (Fig. 1).

Fig. 1 Freeway interchange influence areas
The influence of the mode of operation of such sections on the capacity of highways is due to the process of interaction of the traffic flow leaving the highway (or entering from it) with the transit traffic flow moving along the highway.In order to develop recommendations for determining the capacity of highways, the article proposes a stochastic model of segregation of traffic flows.The particles move in the same direction over a multi-lane cell field.The waste stream occurs at a transit station.There are two types of particles.Particles of the first type move along one strip and are not inclined to move to another.Each particle of the second type tends to move sequentially to the far right band.The length of the section and the intensity of flows on the lanes at the beginning of the section are given.We develop an approach to calculate the probability that a fixed particle of the second type will pass sequentially along the extreme right band on the site.Our approach is based on the version of the flow segregation model developed in [4] (Bugaev A. et al., 2006), [5] (Bugaev A. et al., 2008), and we use the deterministic-stochastic approach developed in [6] (Buslaev A. et al., 2003), [7] (Buslaev A. et al., 2005).We consider a stochastic model of flow segregation.Particles move in the same direction on multi-lane field of cells.Flow departing occurs on a transit section.There are two types of particles.Particles of the first type move along the same lane and do not tend to transit to another lane.Each particle of the second type tends to transit successively to the utmost right lane.The length of the section and the intensities of flows on the lanes in the section beginning are given.We develop an approach to calculate the probability that a fixed particle of the second type will transit successively to the utmost right lane on the section.

Description of the model
Assume that particles of two types move on -lane section.The section is developed into segments.Each segments contains cells.The size (length) of each cell, located on the segment , equals , .The coordinates of a cell are , where is the index of the lane, and is the index of the point.The -th segment of the th lane contains the cells: There are two types of particles.Particles of the first type move along the same lane and do not tend to transit to another lane.Each particle of the second type tends to transit successively to the Kth lane.The particle of the first type correspond to cars that do not tend to change lanes.The particles of the second type correspond to particles that intend to move along another road after passing the section.The value of corresponds to the dynamical dimension of cars.The rule of particles movement on the segment , .(1) If the particle of the first type is on the lane , then it is on this lane while it moves along the section.
For a time quantum , the particle of the first type, located on the lane and the segment with probability tends to move onto one cell in the direction of movement.The attempt is realized if the cell ahead is vacant.
(2) If the particle of the second type is in the cell , and the cell (a cell on the lane to the right of the particle) is vacant, then the particle comes to the cell with probability .
(3) If a particle of the second type is on the lane , and the cell, located ahead this particle is vacant, and the neighboring cell on the lane k, is occupied, then the particle does not move.
Assume that, at time , the particle of the second type is on the m-th segment, on the -th lane in the cell -1.Suppose is the probability that this particle will be in the cell

Calculation of traffic flow characteristic
Let's make the following assumptions: in each cell located on the -th segment on the -th lane, with probability there is a particle of type , With probability , where , this cell is free.Probabilities of states do not depend on each other.Formula for Suppose (k,m) is the density of flow on the lane k of the segment m. Assume that Flow densities for the first segment, i.e. the values of  i (k,1) are specified.Let is the probability that a particle of the second type that started moving along the -th segment on the -th lane, will have time before the end of the passage of this segment to move to the adjacent right lane.The values of of the intensities of the flow of particles moving along the -th lane on the first segment are calculated by the formula q 1 (k,1)(v 1 +(1-r(k,1)) , ∆t), k=1,…,K.

Conclusion
An approach for calculating the characteristics of traffic flow segregation based on the stochastic segregation model is developed.Deterministic-stochastic approach to traffic flow analysis is used.The presented studies can be used to determine the capacity of traffic segregation sections, which will improve the validity of design decisions in the design of highways, which in turn will contribute to improving the quality of transport services.