02.12.202027.05.2017

UDC 656.135

METHODOLOGY FOR ASSESSING PASSENGER FLOW IN URBAN ELECTRIC PASSENGER TRANSPORT

Irkutsk State Technical University, 664074, Russia, Irkutsk, st. Lermontov, 83.

The article presents a methodology for assessing the inter-stop matrix of passenger correspondence according to the data of incoming and outgoing passengers at stopping points (OP) of urban electric transport. The methodology allows to quantify the demand for travel between public transport stops, to analyze existing routes from the standpoint of the efficiency of rolling stock operation and to propose solutions to improve its operation. Il. 5. Tab. 2. Bibliography. 8 titles

Key words: demand for travel; passenger traffic; urban electric passenger transport; travel matrix; interstop matrix of passenger correspondence; O-D matrix estimation.

EVALUATION METHODOLOGY FOR PASSENGER TRAFFIC ON URBAN ELECTRIC PASSENGER TRANSPORT R.Yu. Lagerev, S. Yu. Lagerev, S.S. Nemchinov

Irkutsk State Technical University, 83 Lermontov St., Irkutsk, 664074, Russia.

The article introduces an evaluation methodology for inter-stop passenger correspondence matrix by the data on passengers who get in or get off urban electric transport at public transport stops. The methodology allows a quantitative estimation of the demand for travel between the stops of public transport, the analysis of the existing routes from the point of view of rolling-stock operation efficiency as well as the solutions to improve its operation. 5 figures. 2 tables. 8 sources.

Key words: travel demand; passenger traffic; urban electric passenger transport; trip matrix; inter-stop matrix of passenger correspondences; OD matrix estimation.

When preparing projects for the operation of urban passenger transport, first of all, data is needed that characterize the size and direction of existing or prospective passenger traffic. As you know, such information is graphically presented in the form of cartograms of passenger traffic on the public transport network or in the form of tables of inter-stop passenger correspondence on certain sections of the road network (Table 1).

Long-term foreign and domestic experience of transport interns makes it possible to single out tables of passenger correspondence to the most objective indicators of the load on the public transport network. With the improvement of technical devices for accounting for transported passengers (the introduction of a turnstile system for counting incoming and outgoing passengers, electronic tickets, including contactless payment systems), methods for their assessment continue to develop. The methods are based on the solution of the classical problem of determining

interstop passenger correspondence matrices based on the data of incoming and outgoing passengers at public transport stops, widely used as input data in transport planning and modeling of urban areas.

Tables of inter-stop correspondence on the route determine the required carrying capacity of the route and, accordingly, allow assigning the required number of rolling stock. One can note their common property: they are all characterized by a laborious stage of collecting information and require the involvement of a large number of accountants in the survey. At the same time, currently existing models based on a limited amount of data (gravity models) can only give an approximate idea of \u200b\u200bthe existing passenger traffic in public transport.

1Lagerev Roman Yurievich, Associate Professor of the Department of Management and Logistics in Transport, tel .: 89500697698, e-mail: [email protected]

Lagerev Roman, Associate Professor of the Department of Transport Management and Logistics, tel .: B95GG69769B, e-mail: [email protected]

2Lagerev Sergey Yurievich, Associate Professor of the Department of Management and Logistics in Transport, tel. : 795GG697596, e-mail: [email protected]

Lagerev Sergey, Associate Professor of the Department of Transport Management and Logistics, tel .: 795GG697596, e-mail: [email protected]

3Nemchinov Sergey Sergeevich, Master student of the Department of Electric Drive and Electric Transport, tel .: B9G256B67G2, e-mail: [email protected]

Nemchinov Sergey, Graduate Student of the Department of Electric Drive and Electric Transport, tel .: B9G256B67G2, e-mail: [email protected]

Previously, it was believed that the abundance of factors influencing the formation of transport links does not provide opportunities for their accurate comprehensive accounting. Recently, at stopping points of passenger transport and in public transport in Moscow, automated accounting systems for transported passengers have been introduced, based on counting the number of incoming / outgoing passengers at stopping points, which make it possible to develop sufficiently accurate and reliable methods for predicting inter-object movements and performing their distribution over the network public transport.

Automatic control over the filling of the rolling stock is the most advanced method of passive registration of passenger flows. Recently, much attention has been paid to this type of control, since it allows you to receive data on passenger traffic continuously, quickly and with minimal labor costs. In our country, the most active in this direction is ZAO NPP Transnavigation, which has developed a software and hardware complex called ASM-PP (Automated Passenger Traffic Monitoring System).

The main purpose of the ASM-PP is to keep track of incoming / outgoing passengers, collect information about the filling of the cabin, determine the level of actual demand for transportation, and actually record production flights. In addition to the contact-turnstile approach, until now, there have been practically no other options for automatic accounting of passenger traffic on the domestic market.

Thus, the assessment of the quality of the functioning of any transport network is closely related to the structure of passenger movement between stopping points. Therefore, the calculation of the value of interstop passenger movements can be attributed to the central task of accounting and formation of passenger flows on any network of passenger transport (tram and trolleybus lines, metro lines). The main quantitative characteristic of the structure of passenger movement over the network is the passenger exchange table, the elements of which are the volumes of passengers per unit of time between each pair of stopping points (Table 1).

Tables of inter-stop passenger traffic on public transport remain one of the main means of quantitative analysis in transport design and serve as a first approximation for analyzing the size and structure of inter-district urban and suburban communications, as well as the basis for solving problems of choosing express and shortened routes and justifying the choice of a bus schedule and trains on suburban sections.

Currently, in many Russian cities, a large number of studies of passenger traffic are based on the use of detectors that allow collecting detailed data on passengers, including in real time. At the same time, most of such studies are still in progress.

manually, using the accountants. Such surveys are carried out to clarify traffic plans, redistribute rolling stock along routes and hours of the day, clarify the route system, and resolve issues of transport coordination. The types and methods of field surveys of passenger traffic on the routes are well and in detail covered in special literature and relevant manuals.

Table 1

General view of the interstop table

passenger mail

Input stop number Output stop number

1 a 0 X12 X13 ... X1p

2 a2 0 X23 ... X2p

3 a3 0 ... X3p

The authors of this article present the results of a mathematical algorithm for evaluating tables of interstop correspondence, based on solving a linear programming problem, when the initial data on incoming and outgoing passengers on the route may contain calculation errors.

In this case, the task is aimed at finding the elements of the table. 1 xy, characterizing the number of passengers who passed between i and j stopping points, xy\u003e 0, using the data of counting the number of incoming / outgoing passengers at each of the stopping points of passenger transport. The sum of the elements of the i row of the matrix corresponds to the number of passengers entering the i OP, and the sum of the elements of the j column of the matrix corresponds to the number of passengers who entered the j OP:

a \u003d X xy; b \u003d X xy; ^ \u003d 1 ..... ^ (1)

while a | and b satisfy the condition

The first and natural step towards solving this problem is an attempt to establish a quantitative relationship between the values \u200b\u200bof inter-stop movements and the filling of cars (rolling stock). A similar problem arises in computed tomography, when it is necessary to reconstruct the object itself from a certain set of available projections of an object.

In matrix form, the problem of evaluating the table of passenger correspondence is presented in Fig. 1, where it is necessary according to the given intensity values

of movements y through the routing matrix A determine interstop flows x ..

The estimation problem consists in finding such values \u200b\u200bof the correspondence vector x for which the calculated values \u200b\u200bof the rolling stock filling on the arcs of the network graph y (y \u003d Ax) coincide as much as possible

with observed y values:

If Cr_x\u003e 0; if Cr j \u003d 0,

Z Kl \u003d Ë | Y - Yr \\ ^ min.

Consider an artificial tram route (Fig. 2).

For the matrix M, a solution is obtained in the following form:

At the same time, it should be noted that in most situations encountered in practice, the number of arcs for which there are sufficiently reliable data on flows (values \u200b\u200bof incoming / outgoing passengers, filling of rolling stock on the tracks) is significantly less than the number of corresponding pairs of stopping points (values \u200b\u200bof passenger exchange x¡ j). This means that the number of unknowns in the system significantly exceeds the number of equations and, therefore, the above systems may be incompatible.

Figure: 1. Representation in matrix form of the problem of estimating matrices of interstop passenger correspondence

Fig 2. Representation of the tram route in the form of a graph (arrows indicate the number of incoming and outgoing passengers at stopping points)

In this case, the traditional way of obtaining solutions is the construction of special mathematical programming problems, in which the discrepancies between the projections of the calculated values \u200b\u200bof the corresponding passenger flows and the given ones are minimized. According to this principle, the authors of the article have developed a mathematical algorithm for calculating the inter-stop passenger traffic according to the data of incoming / outgoing passengers, based on linear programming algorithms.

As a basis for the proposed methodology for evaluating passenger tables between corresponding OP

if C ^ j\u003e 0;

if C_j \u003d 0;

table 2

The resulting table of interstop correspondences of the tram route (see Fig. 2)

Arrival stop number Total

0 1 2 3 4 5 6 7 8 9 included, pass.

and; and 0 0 3 4 4 1 3 2 4 2 2 25

w s; m th Œ 1 0 1 1 0 1 1 1 1 1 7

2 0 1 0 1 1 1 1 1 6

1- o 3 0 1 2 2 3 2 2 12

^ m 4 0 2 1 2 1 1 7

th b 5 0 1 2 1 1 5

o o cf 6 0 2 1 1 4

Total left, pass. 0 3 5 6 2 9 8 15 10 10

the method of least modules is chosen, reduced to a linear programming problem with linear and two-sided constraints. A solution is proposed for finding the matrix of correspondences in the form

under linear constraints on the variables ^ 2x2 \u003d y, x2\u003e 0, and two-sided constraints

xlb< x2 < xub, где xlb и xub - векторы нижних и верхних ограничений оцениваемых параметров, xlb < 0, xub > 0. Here, the components of the vector x2 are the estimated values \u200b\u200bof passenger flows between each pair of OP (j \u003d 1 ..... m) and the convergence errors of the turnstiles data (incoming / outgoing passengers, passenger compartment filling on the tracks) with the projection data of the estimated passenger traffic table (j \u003d m + 1, ..., m + 2n), m is the number of corresponding stopping points, n is the number of route graph edges on which the passenger traffic values \u200b\u200bare known, A2 is the transformed incidence matrix A, y is the vector of known passenger traffic values \u200b\u200b(turnstile data).

Incident matrix A, i.e. the matrix characterizing the belonging of interstop correspondences to arcs of the route graph will have the structure shown in Fig. 3.

As an approbation of the presented technique, let us consider an artificial train route with the initial "0" and the final "9" points (see Fig. 1). The initial data are the values \u200b\u200bof incoming / outgoing passengers at each stopping point and, therefore, the amount of filling of the rolling stock at each of the 9 stages.

The computational procedure for finding the vector of values \u200b\u200bx2 is an iterative process, at each step of which the projection errors between the calculated values \u200b\u200bof the interstices are minimized.

new table of passenger traffic with data from turnstiles located at passenger stations (Fig. 4).

5 10 15 20 25 30 35 40 Corresponding OP

Figure: 3. The structure of the matrix A for the graph of the route under consideration

Iteration number

Figure: 4. Convergence of experimental values \u200b\u200bwith calculated ones (obtained as a result of overlaying the passenger exchange table on the route network)

10 15 20 25 30 Detector data, pass.

0 5 10 15 20 25 30 35 40

Detector data, pass.

Figure: 5. Errors of convergence of values \u200b\u200bof passenger traffic (experimental and calculated data) at the 3rd iteration

On the whole, testing of the methodology using the data example (see Fig. 2) showed its rapid convergence, a sign of which is the appearance of horizontal sections on the graph (see Fig. 3). In this case, convergence is achieved already at the 3rd iteration. The average absolute error of 1.55 obtained at the 3rd iteration (see Fig. 3) and the ratio of the average absolute error to

the average value of passenger traffic allows us to assert that this approximation is valid. The resulting structure of passenger traffic between the corresponding OP of the route under consideration is presented in table. 2.

The correlation coefficient between the values \u200b\u200bobtained as a result of the entry / exit calculations and the projections of the passenger exchange table on the route network reaches 0.97, which confirms the high quality of the regression (Fig. 4).

Based on the results of testing using an artificial tram route, it was found that the method has good convergence, is efficient, and is effectively used for matrices of incomplete rank. This allows it to be used to inspect the movements of passengers between stopping points of urban passenger electric transport, using the automated systems for recording incoming / outgoing passengers that have been introduced recently, which allows calculating the entire set of necessary route characteristics: the number of passengers carried, the filling of the cabin along the route length, uneven passenger traffic by time and directions (forward and backward), average trip length, etc. The availability of this information in real time will significantly improve the quality of the operational management of the GEPT.

Bibliographic list

1. Artynov A.P., Skaletsky I.I. Automation of planning and management of transport systems. Moscow: Transport, 1981.280 p.

2. Zedgenizov A.V., Lagerev R.Yu. Influence of the traffic light operating mode on the throughput of stopping points // Izvestiya vuzov. Investments. Construction. The property. 2011. No. 1 (1). S. 38-44.

3. Mikhailov A.Yu., Golovnykh I.M. Modern trends in the design and reconstruction of street and road networks. Novosibirsk: Nauka, 2004.266 p.

4. Lagerev R.Yu. Calculation of correspondence matrices of traffic flows using an algorithm that is resistant to errors in the initial data // Bulletin of the Irkutsk State

the state technical university. 2007. No. 1 (29). S. 161164.

5. Levit B.Yu., Livshits V.N. Non-linear transport systems. Moscow: Transport, 1972.144 p.

6. VN Myagkov, NS Palchikov, VP Fedorov. Mathematical support of urban planning. L .: Nauka, 1989.144 p.

7. Lam W.H.K., Lo H.P., Zhang N. Estimation of an origin-destination matrix with random link choice proportions: a statistical approach // Transportation Rese., 1996.30B. P. 309-324.

8. Nihan, N.L., and G.A. Davis. Recursive Estimation of Origin-Destination Matrices from Input / Output Counts // Transportation Research-B, 1987. Vol. 21B. N2. P. 149-163.

The movement of passengers in one direction of the route is called passenger traffic. Passenger traffic can be in the forward direction and in the opposite direction.

A characteristic feature of passenger flows is their unevenness, they change over time (hours, days, days of the week, seasons).

The passenger traffic is characterized by:

Power or intensity, that is, the number of passengers passing at a certain time on a given section of the route in one direction;

The volume of passenger traffic, that is, the number of passengers carried by the considered mode of transport for a certain period of time (hour, day, month, year).

The distribution of passenger traffic on the route (by hours of the day and route sections) are presented in Table 1 and Table 2.

Table 1

Hours of the day	Number of passengers	Hours of the day	Number of passengers
Direct direction	Reverse direction	Direct direction	Reverse direction
6-7			16-17
7-8			17-18
8-9			18-19
9-10			19-20
10-11			20-21
11-12			21-22
12-13			22-23
13-14			23-24	-	-

Settlement and technological section

Characteristics of passenger traffic

Passenger traffic is the number of passengers that are actually transported at a given time on each leg of the bus route or, in general, on the bus network of all routes in one direction per unit of time.

As a rule, passenger flows are not the same in size at different hours of the day, days of the week, months and seasons of the year, as well as by sections of routes and directions of bus movement.

In order to identify passenger flows, distribute them by directions, collect data on changes in passenger flows over time, surveys are carried out. Objective of the survey: to obtain reliable data on the capacity, distribution and fluctuations of passenger traffic on bus routes.

Passenger flows are depicted in the form of graphs, cartograms, diagrams, or recorded in tables.

Survey methods are classified according to a number of characteristics:

By the duration of the period covered:

Systematic (daily, weekly, etc.);

One-time (short-term);

By width of coverage:

Continuous (simultaneously throughout the entire transport network of the served area), on average, once every 3 years;

Selective (for individual traffic areas) 1 time per quarter;

By type:

Questionnaire method (by filling in pre-developed special questionnaires);

The reporting and statistical method is based on ticket-registration sheets and the number of tickets sold;

Coupon method (by issuing specially prepared coupons of different colors to the accountants);

Tabular method (carried out by the accountants located inside the bus near each door, by filling in pre-prepared tables);

The eye method (carried out by collecting data on routes with significant passenger traffic, is carried out visually according to a point system from 1 to 5 points). It can be used by drivers or conductors.

Silhouette method - a kind of visual (5-point system, by a set of silhouettes by bus type);

Poll method - by polling the passenger in the cabin by the accounting officer, this method allows you to determine the data on the correspondence of passengers.

Passenger traffic by hours of the day and route sections (forward and backward directions) are presented in Table 3, Table 4, Table 5.

Table 3

Hours of the day	Passengers carried
Direct direction	Reverse direction	In both directions
6-7
7-8
8-9
9-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
18-19
19-20
20-21
21-22
22-23
23-24	-	-	-
24-1	-	-	-
Total

Table 4

Direct direction

Table 5

Reverse direction

The study of passenger flows on individual routes is carried out in order to improve the quality of transport services for passengers, for which they use information obtained in solving technological problems of the selection and distribution of rolling stock, rationalization of modes and timetables. The intensity of passenger traffic is presented in tabular form or graphically for different days of the week, periods of the day, directions of bus movement on the route. The volume of traffic along the route sections is determined by one of the experimental methods:

Eye method

Silhouette method

Weight method

Tabular method

Automated method

Poll method

Coupon method

Calculation and analytical method

The distribution of passenger traffic on the route (by hours of the day and route sections) are presented in Table 1 and Table 2.

Table 1 - Distribution of passenger traffic by hours of the day

Hours of the day	Number of passengers
Direct direction	Reverse direction

Table 2 - Distribution of passenger traffic by route sections

Characteristics of passenger traffic

Passenger traffic - the number of passengers actually passing in a given direction at a given time.

Methods for studying passenger traffic:

The questionnaire method of surveying passenger traffic is based on filling out a survey questionnaire by citizens or special accountants about the number of trips, the purpose and methods of movement, routes, places of transfer, time of movement and to clarify other issues, depending on the purpose of the survey.

Reporting and statistical methods are based on the use of data from the current system of accounting and reporting on transportation. Practical application is limited by the availability of accounting indicators. The method is the main one for surveys carried out on intercity and international routes. In cities, the reporting and statistical method provides information on the total volume of passenger traffic, the distribution of hourly revenue of taxi cars.

Experimental methods are based on surveys carried out according to the developed programs and methods. This method is the main one for survey of entrances and exits on intracity and suburban routes.

Computational and analytical methods are based on the use of passenger formation and passenger absorption models, forecasting models of indicators characterizing the needs for transportation. Scope - clarification and correction of data obtained during surveys.

The eye method is used by the bus driver, who is given a special uniform before entering the line. Being on the busiest stretch of the route, the driver assesses the filling of the bus with passengers by eye. The silhouette method is used when examining the filling of buses at bus stops. Past

preliminary preparation, the accountants visually assess the filling of the bus "in the light". During the examination, the clerk writes down in the form the time of passage of a bus of a certain type with filling corresponding to the most similar silhouette.

The weighting method for calculating the passenger compartment filling involves the use of sensors mounted on the bus air springs. The sensors generate signals proportional to the mass of passengers on the bus. The average passenger weight is 70 kg.

The tabular method is used in two versions - the survey is carried out on the bus or at bus stops. During the examination in buses, the accountants are located near the doors and fill out a special tabular form. At each stopping point, a clerk counts the number of passengers who have exited and entered, and makes a corresponding entry in the form.

The automated method reduces the labor intensity and cost of examining passenger traffic. The recorder consists of sensors that record the entry and exit of passengers, a data recording unit and a power supply unit connected to the on-board electrical network of the bus. The determination of the direction of movement of passengers (entrance or exit) is provided by the logical device of the registration unit.

The survey method is provided by clerks located in buses at each door. They interrogate the passengers who have entered: to what stop they are going, whether they make transfers during the trip and on which routes. The data is entered into the table.

The coupon method is the main one in determining inter-stop correspondence. When boarding, each passenger is given a special ticket, which is handed over to the accounting officer upon exiting the bus. Calculation - analytical method allows to rationalize the volume of the survey of correspondence.

A reliable picture of the correspondence is provided by the survey of only all bus routes in operation.

On route 49, the passenger traffic was surveyed by a visual method.

Table 3 - Passenger traffic by hours of the day

Hours of the day	Passengers carried
Direct direction	Reverse direction	In both directions

Table 4 - Passenger traffic by route sections

Direct direction

Table 5 - Passenger traffic by route sections

Reverse direction

2.1.1 Determination of passenger traffic

P \u003d Ppr + Rob (pass-km), (2.1)

where Рпр - passenger turnover in the forward direction, passenger-km

Rob - reverse passenger turnover, passenger-km

P \u003d 27935+ 26501 \u003d 54436 passenger-km

Passenger transportation is one of the most important sectors of the economy. One of the main problems of passenger transport enterprises is the shortfall in receiving significant amounts of income, which ultimately affects the quality of transportation and passenger service. Practically all enterprises are faced with the issue of improving work efficiency and saving resources, so accounting for the number of passengers is becoming one of the most important tasks.

The advantages of CAC

The automatic passenger counting system (CAS) can benefit both bus and rail operators and passengers: the former have the ability to match the number of passengers with payment information to successfully combat free riders, and the latter can receive real-time information on the location of the transport means (vehicle) on the route and the availability of free seats in it.

In addition, the capabilities of the system can be expanded for planning maintenance, taking into account the intensity of the use of transport, which will optimize the load of the vehicle by region or time period, as well as control the quality of service.

The following applications can be deployed using such systems:

optimization of vehicle availability;
identification of "hares" in public transport;
increasing the efficiency of vehicle utilization through fleet management and maintenance planning;
limiting the number of people to ensure safety;
analysis of traffic flows in cities.

The diagram (Fig. 1) demonstrates the variant of the passenger SAS operation. The sensors are installed in regional trains on the Venice – Belluno route. They count the number of people entering and leaving at each station and, tied to the payment system, allow identifying the presence of stowaways.

Figure: 1. An example of building a report in the passenger counting system

The issues of accounting for the number of passengers are relevant not only for urban public transport, but also for suburban trains and metro. In addition to such an important task as catching rabbits, SAS help to optimize the train schedule in various directions, since they can be used to collect statistics on how many passengers, at what stations, and at what time they board and disembark. These systems can improve the service provided not only on the transport itself, but also at airports, train stations, etc. e. Enumerators can control, for example, the number of people queuing up for passport control in order to vary the number of working checkpoints, if necessary; information about waiting to board the plane allows you to optimize the movement of buses from the terminal to the plane. Accounting for the number of passengers can also be useful for river / sea transport, in order, for example, to avoid overloading ferries and motor ships, which can lead to tragic consequences.

Problems and solutions

The introduction of new technologies improves the operation of the transport system. Counting passengers has been challenging enough in the past. Initially, the drivers had to manually
counting the number of passengers, which was a rather difficult task and did not give accurate results. Despite the development of sensor technologies (light-sensitive, infrared, heat sensors), the systems built on their basis still fail to accurately calculate the passenger traffic. Passengers can be of different heights, be in a wheelchair, move at different speeds, records can be kept in different weather conditions during the day. Most sensors in such conditions are not able to provide reliable operation. For more efficient counting, new technologies are required that can operate in harsh conditions, under various lighting conditions and conduct accurate counting when a large flow of people passes at the same time. One such technology used in Eurotech sensors is stereoscopic vision.

Stereoscopic vision technology

Figure: 2. People counter DynaPCN 10-20 by Eurotech

The Eurotech DynaPCN 10-20 counter (Fig. 2) is a compact device based on non-contact stereoscopic vision technology specially designed to count the number of passengers entering / leaving trains, buses, subway cars, etc.

Naturally, the most important characteristic for such devices is the counting accuracy. A well-designed sensor should be intelligent and flexible to accommodate differences in passenger heights or multiple passengers crossing a bill border at the same time. DynaPCN 10-20 achieves high accuracy thanks to the use of two stereoscopic cameras and four high-brightness infrared LEDs.

Stereoscopic cameras capture an image under the sensor and process this data in real time using a special sophisticated algorithm. The algorithm analyzes the height, shape of any objects falling into the field of view and the direction of their movement. It accurately identifies a person by characteristic body parts - it is enough to highlight only the head or shoulders to recognize a person with a minimum level of error. When the sensor determines the direction of a person's movement, the corresponding counters are incremented, while storing information about the current time. A time slice of the intensity of the passenger flow will make it possible to optimally plan traffic routes depending on the workload at specific time intervals. Thanks to the use of advanced passenger counting technologies, the DynaPCN 10-20 achieves an accuracy of 97% in real-world operation.

Flexible installation

The DynaPCN construction allows for easy and inconspicuous flush mounting in the ceiling space above the doorway and can be adapted to doors and roofs of various designs. The angle of the optical panel can be changed, which makes it possible to place the sensor in various places, including on non-horizontal surfaces. The presence of specialized digital I / O channels for connection to door status sensors and easy integration of several sensors into one counting system simplify installation in any type of transport.

The Eurotech CAC is constantly evolving. It has become more flexible in terms of meter installation height for wide doors. For doors up to 1.8 m wide, only one DynaPCN is now required, installed at a height of about 2 m.Previous versions of the meters required the installation of two sensors for such wide doors, but the updated software allows you to capture images from two cameras in one device and build on it based 3D model.

If the size of the doors exceeds the area monitored by the sensor, you can install two or more sensors by defining in the settings of each device its own dead zone in the overlapping areas to ensure correct counting (Fig. 3).

Figure: 3. An example of using sensors for narrow and wide doors

Protected execution

Passenger counters must operate over a wide range of different conditions and environmental influences. Different lighting conditions and other external factors can lead to counting errors with many solutions currently available on the market. The DynaPCN meter is specially designed for use in transport and has been tested over a wide temperature range and when exposed to vibration. Thus, this solution can be applied to work in harsh operating conditions. DynaPCN uses an integrated LED infrared illumination to achieve high accuracy under various lighting conditions
high brightness, which allows for accurate passenger counting in any light, even in the dark.

DynaPCN offers a set of discrete I / Os with which it can communicate directly with smart doors or motion control systems, thus guaranteeing the best counting performance. For example, discrete inputs can be connected to door open / close sensors, which will enable / disable counting under appropriate conditions.

Meter connection using Power-over-Ethernet (PoE)

DynaPCN connects to the on-board computer via an RS-485 interface, and as technologies such as the use of infotainment services develop in vehicles, it switches to Ethernet. Using a standard Ethernet connection, the DynaPCN 10-20 can be easily integrated into existing vehicle networks. Power-over-Ethernet (PoE) technology is also supported, where power is supplied over an Ethernet cable, making it easy to install the DynaPCN in a vehicle. The ability to configure IP addresses allows you to organize remote control and update the sensor software. Optional RS-485 support allows DynaPCN to be used not only in transport, but also in other industries, such as retail.

Work in the system

Each sensor is a complete solution. It is enough to install it and connect it to the on-board computer via RS-485 or Ethernet, configure it, and it is ready to go. With a simple set of commands, information on counting events can be read from the sensors. The counters have a built-in memory that can store more than 1 million account events. There are a number of scenarios for working with the sensor:

Other uses for SAS

An alternative use case for DynaPCN is to use it to count the number of store visitors. The sensor can be connected to the Eurotech Everyware Device Cloud for quick access to data. The customer can correlate the number of people entering various stores (within the chain) with the number of purchases made to determine the performance of the various stores and the problems associated with their location. In addition, depending on the number of people in the room, it is possible to adjust the operation of the ventilation system, lighting, etc.

Another use case for DynaPCN is to organize toilet cleaning based on the number of visitors. Cleaning of toilets in large buildings is usually based on temporary ‘ x cycles calculated according to standards. These standards also determine the number of personnel required to perform work on a given schedule. This approach has a number of pitfalls:

Actual use of toilets is not included. This means that the most visited ones are removed at the same frequency as the less visited ones. Accordingly, some are cleaned unnecessarily often, while others are not cleaned properly.
The calculation does not take into account the time of day (the interval between cleanings is almost constant) and the change in workload depending on the day of the week, which again leads to both the inefficiency of cleaning and the inefficiency of staff employment.
Peak events are not included.

The carried out statistical analysis showed that there is no dependence both on the time of day and on the day of the week. It is very ineffective to build a work schedule only for them. To optimize the work, the people counting system from Eurotech was used. Meters are installed on each door, the data from which are collected in a central computer. When the set value of the past visitors is reached, a command is issued to clean a specific room. This approach, in contrast to standard methods, can significantly improve the quality and efficiency of cleaning work.

Benefit for users: From a visitor's point of view, the quality of the content of objects is significantly improved, during peak periods, toilets are cleaned more often than before, so that they rarely remain dirty.

Benefits for Building Management: The system collects real-time information on the usage patterns of all toilets in the building. Based on this data, it is possible to track the patterns of use and their relationship with other events, which will allow making predictions and optimizing the work schedule of personnel, reorienting employees to other jobs, etc. This also optimizes the consumption of detergents.

The Eurotech passenger counter, based on stereoscopic vision technology, delivers accurate results in a wide variety of environments, including severe operating conditions. When connected to an on-board computer, it becomes a highly efficient SAS, capable of better organizing transport operations, as well as solving other tasks, such as monitoring visitors in shops and public places.

A successful solution to the issues of rational organization of passenger transportation and efficient use of rolling stock is impossible without a systematic study of the nature of changes in passenger traffic in the transport network. The study of passenger traffic allows us to identify their distribution in time, route length and directions of movement. Various methods are used in the study of passenger traffic. Existing methods for surveying passenger traffic can be classified according to a number of criteria.

1. By the duration of the covered perioddistinguish between:
- - systematic surveys;
- - one-time examinations.

Systematicsurveys are carried out daily during the entire period of movement of vehicles along the route, as a rule, by employees of the operation service of passenger transport enterprises.

One-timesurveys are short-term surveys carried out within the framework of a developed program, determined by the goals set: opening or closing a route, determining the capacity and required amount of rolling stock, etc.

2. By the width of the coverage of the transport networkdistinguish between:
- - continuous surveys;
- - sample surveys.

Solidsurveys are carried out simultaneously throughout the entire transport network of the served area. They require the involvement of a large number of workers (accountants). Based on the results of continuous surveys, global issues are resolved: the efficiency of the transport network, the directions of its development, coordination of the work of various types of transport, changing the route scheme, choosing modes of transport in accordance with the capacity of passenger flows, etc.

Selectedsurveys are carried out in separate areas of the route network, conflict points or some routes in order to solve local, private, narrower and specific problems.

3. By the wayallocate:
- - questionnaire surveys;
- - reporting and statistical surveys;
- - field surveys;
- - automated examinations.

Questionnaire method,as a rule, it covers the entire route network of the served area and allows you to identify passenger flows for all types of transport. It is characterized by continuous examination. The questionnaire method allows you to establish the potential mobility of the population: the real needs for movement in terms of quantity and directions, regardless of the existing route network. This method provides for obtaining the necessary information using pre-developed special survey questionnaires. The success of the questionnaire survey and the reliability of the data obtained are largely determined by the nature, simplicity and clarity of the questions posed. Therefore, the form of the questionnaire must be carefully thought out in accordance with the set goal and provide for the possibility of its machine processing. The survey is carried out in places of mass concentration of the population. The questionnaire survey gives the greatest effect when polling at the place of work of the population: at the main passenger-forming and passenger-absorbing points of the served area. In this case, employees of organizations (employees of the personnel department) can be involved in the survey. The complexity of this survey method lies in the processing of questionnaires. In order to reduce the complexity of processing, the questions and answers of the questionnaire can be encoded and then processed using a computer.

Reporting and statistical methodthe survey relies on data from ticket-registration sheets and the number of tickets sold. In addition to the tickets sold, it is necessary to take into account the number of persons transported on monthly travel tickets, service certificates, persons enjoying the right of free reduced travel, as well as those who have not purchased a ticket. Using the reporting data, it is possible to determine the traffic volumes on individual routes, establish the distribution of passenger traffic by hours of the day, days of the week, etc. But this method does not allow assessing the distribution of passenger traffic by route segments, that is, to establish the maximum load of rolling stock on the route.

Field surveysimply obtaining information about the actual movements of the passenger through direct interaction with them. Field surveys can be coupons; tabular; visual; silhouette; polls.

Coupon methodsurveys of passenger traffic allows you to establish information about the capacity of passenger traffic along the length of the route and time of day, on the passenger exchange of stopping points, correspondent communications, the average distance of a passenger's trip, filling the rolling stock, etc. ... In the course of the survey, the accountants at each stop of the route issue coupons to all passengers entering the passenger compartment of the vehicle, having previously noted the number of the stop at which the passenger entered. For each direction of travel, their own coupons are used, as a rule, of a different color, with ascending or descending stop numbers. When leaving the vehicle, passengers hand over the coupons, and the accountants note the number of the stop at which the passenger left. If a passenger makes a transfer, he makes a corresponding mark on the ticket (tears off the spine). At the final stop, the accountants hand over the used tickets for a specific flight to the controller and receive new ones. To conduct a survey by this method, preliminary training is required, which includes the development of a program and the calculation of the required number of accountants and controllers. The survey program determines the technological sequence of the work with the indication of terms. The quality of the information received largely depends on the accuracy of the work of the accountants and controllers, as well as on the preparedness and awareness of passengers.

Tabular methodsurveys are carried out by surveyors who are also located inside the vehicle near each door. Counters are supplied with survey tables, which indicate general information about the vehicle, flight number, departure time, route stopping points for each direction. For each stopping point of the flight, the accountants enter in the appropriate columns the number of passengers who entered and left, and then calculate the filling in the sections between the stopping points of the route. Passenger registration is carried out by each accountant separately, and the processing of the received data is carried out jointly. The tabular method can be used for systematic and one-time, continuous and sample surveys. In case of continuous and systematic surveys, the form of tables should allow the processing of survey data using a computer.

Visual (eye) methodsurveys are used to collect data at stopping points with significant passenger traffic. Counters visually determine the filling of the vehicle according to the conditional point system, and this information is entered into tables. For example, 1 point is awarded when the vehicle has free seating; 2 points - when all the seats are occupied; 3 points - when passengers stand freely in the aisles and storage areas; 4 points - when the nominal capacity is fully used and 5 points - when the vehicle is overcrowded and some of the passengers remain at the stop. Points are entered in the table according to the make and model of the vehicle. Knowing the capacity of a particular brand and model, you can go from points to the number of passengers carried. With the help of this method, data on the occupancy of the rolling stock can be obtained by sections of the route, but it does not allow us to establish the real volume of passengers transported along the route in general and the nature of correspondence. Visual inspection can be carried out by drivers or conductors, who are given a corresponding table. At the end of the shift, the tables are handed over to the line dispatchers, and in the operation department they are processed and the number of passengers who traveled along the routes and sections is determined. This method is mainly used for sample surveys.

Silhouette methodsimilar to the visual method. Only instead of a point assessment of the filling of vehicles, a set of silhouettes by types of rolling stock is used. The counting officers select the silhouette number that matches the filling of the vehicle and mark it in the table. Each silhouette corresponds to a certain number of passengers. Based on the collected silhouette data, the number of passengers in the cabin is calculated when the vehicle moves along the route section.

Poll methodpassenger traffic surveys propose the use of accounting officers who, while in the passenger compartment, ask incoming passengers about the exit point, transfers, the purpose of the trip and record this information. The survey method refers to field surveys and differs from the questionnaire survey because the survey is conducted only among direct users of passenger transport. This method allows obtaining data on passenger correspondence, which helps to correct routes and develop organizational measures to reduce travel time and reduce passenger transfers.

Automated methodsensure the receipt of information about passenger traffic in a processed form without involving people in the direct collection of such information. There are several methods for automated survey of passenger traffic, in particular, contact; non-contact; indirect; combined.

Contact methodsallow receiving data on passenger traffic through the direct impact of passengers on technical means. One way to obtain information could be through the use of automatic devices with a screen and keyboard. Potential passengers (residents of the settlement, visitors, etc.) enter information about the needs for movement into the automatic device by pressing the appropriate keys. The devices can be placed in passenger-forming and passenger-absorbing nodes (train stations, shopping centers, etc.), as well as at stopping points. This method of surveys allows you to obtain information about the correspondence of passengers, the mobility of the population and conduct a sociological survey on the level of satisfaction of the population with the work of transport, etc. The obtained information can be used to optimize the route scheme, change the timetables, etc.

Non-contact methodsuse photovoltaic devices. For photoelectric metering of transported passengers, photoconverters are used, which are installed in doorways or on the outer side of the vehicle, two for each flow of embarkation and disembarkation of passengers. When entering or exiting, passengers cross a beam of light rays coming to photo sensors that record the movement of passengers. Electrical impulses from photosensors enter the decryption unit and, depending on the order of arrival, are sent to the register of incoming and outgoing passengers. The digital display unit sums up the number of passengers entering and leaving at each stop. The disadvantages of this method include the complexity of setting up and adjusting photoelectric sensors, large inaccuracies (up to 25%) during peak hours.

Indirect methodaccounting for passengers carried involves the use of special devices that allow weighing all passengers of the vehicle at the same time, followed by dividing the total mass of passengers by the average weight (70 kg). The total mass of passengers is determined using strain gauge transducers located on the spring cushions. Survey data are presented in the form of diagrams of passenger flows by route sections.

Combined methodregistration of passengers involves the joint use of any automated methods at the same time, for example, indirect and non-contact. This increases the completeness and accuracy of the information collected. Automated surveys of passenger traffic provide a constant and continuous receipt of information on traffic volumes at relatively low cost, since there is no need to involve a large number of people and additionally process the collected information.

Figure 9.4 shows a graphical representation of the classification of methods for surveying passenger traffic.

Figure 9.4 - Classification of passenger examination methods