DE19852857A1 - Method and equipment for calculating connection journeys in transport systems e.g., road, rail, air and ferry services linked by time tables, allowing for actual disruptions by linking existing connection planners and disruption managers - Google Patents

Abstract

System (2) employs two data processing units, connection planner (4) and disruption manager (6) operating independently of one another. The planner knows the valid timetable but no actual disruptions. The manager is supplied with system topology (10) and disruption (12) data but not the timetable. Both are linked by coordinating unit (8) which contains information on start and destination points.

Description

The invention relates to a method and a Device for the automatic generation of Traffic information in scheduled traffic systems and to a device for performing the method.

Traffic systems are in schedule-bound and schedule-free Differentiated traffic systems. To the schedule-bound Transport systems include public transport systems with his means of transport bus, tram, subway, S-Bahn, Regional train and ferry as well as long-distance transport networks with various means of transport such as plane, train, tourist bus and ferry. The timetable-free traffic systems show drivers of Cars (private cars, rental cars or taxis), trucks and Excursion buses, how they can use the streets. Traffic information makes it easier for the user Traffic system to use this. The state of the art knows Traffic information systems for both scheduled and  also for timetable-free traffic systems. To such a system always belongs to a facility, the connection information certainly. This connection information informs the Users on how best to get it from its starting point comes to his goal and what he has to consider. Since For a long time, electronic devices are becoming too route (especially motorway, train station, airport) used or the user in the form of DV programs for the PC (off-line by floppy disk or CD-ROM or on-line by Internet) made available. Different car manufacturers offer recently in the vehicle itself devices that Calculate connection information. Currently on the product mobile connection information devices are being developed, which the Carries user as a mobile phone.

To determine connection information, both at least one connection path to get from one place to another Place to get, as well as information about properties of the possible connection paths can be determined. These properties can be both static in nature and a difficult route or long transfer times as well as dynamic nature like traffic jams, Delays and current interruptions. In schedule-bound Traffic systems must always be in addition to the connection routes also connecting journeys can be specified for the user knows when and how to use the proposed connection can. The term connection planner is used below Designates facilities that determine connecting trips and can spend. A connecting trip consists of an or several individual connections. A single connection consists of Start and end stop, departure and arrival time and the line to use. An example of a Single connection: "Take the S5 from the main station at 8.23 a.m. to Ostbahnhof, arrival 8.37 a.m. "The term" stop "is used in the following as a generic term for bus stop, train station, airport  and jetty used. The single connection can, so Definition, without changing from the start to the end stop to be used. The intermediate stops therefore do not need to be specified.

The current state of the art of connection planners is the following: For many public transport companies and Transport associations (hereinafter referred to as transport providers) Connection planners exist that are commercially available off-line Software (e.g. BVG driving information), on the Internet or on specially designed information columns can be used. These connection planners are from the transportation providers either created on your own or from it specialized companies like Hafas or Mentz that sell their product but do not sell off the shelf, but on the respective one Transport providers according to their wishes and special features cut to size.

Use almost all state-of-the-art connection planners only the information of a computer available Timetable. This timetable may become outdated, i. H. not with that current actual range of trips match. In order to the connection planner is only able to do the Connection journey before departure and according to the timetable determine. He is not able to with current faults to calculate the actual ride offer and while driving to update. Examples of such disorders are: Route is not usable at all, but blocked. Or it can only be slower than usual or only with a part of that Vehicles used by traffic providers are used. On The station and all the tracks are completely closed. A train station is blocked in such a way that the trains pass without stopping.  

So that a connection planner can take faults into account, the schedule would have to be adapted to the current disturbance become. This is economical only in the case of such faults possible that remain the same for months, but not for those which are only available for a few hours - and the latter applies to a great deal many of the disturbances of a traffic network. Should he Connection planner the timetables of several transport providers evaluate or take into account such faults that at the same time the means of transport of several transport providers would affect different timetables and not just a single update.

Current disruptions are common in many traffic providers today a data processing system. The Data processing system receives - by manual entry or automatic recording - information about interruptions, Delays and other disabilities such as Capacity restrictions. This information is in the first place Line recorded with the aim of planning and internal processes document the actions taken. Some transport providers make this information public, e.g. B. the BVG and the Deutsche Bahn. Sometimes the internet is used to Provide fault information. A facility that thus stores information about faults and on treated, is referred to below as a fault manager.

In [Hans-Peter Schär: Munich Comfort - The first on-line and real-time Information System, Proceedings 4 ^th World Congress on Intelligent Transport Systems, October 21-24, 1997, Berlin] the system EFA Realtime is described, in which the connection planner takes into account entered faults and makes trouble-free travel suggestions. The work mentioned does not disclose how the connection planner works. However, there are Internet pages in which the EFA core is described: From [http://www.mentzdv.de/EFA/efakern.htm] it emerges that the current faults are taken into account by the fact that the The connection planner's underlying schedule for the time of the fault is automatically changed, that is to say is adapted to the fault. The connection planner then works unchanged.

EFA and any other traffic information system according to the state the technology uses its own connection planner - this applies also for traffic information systems like EFA Take disruptions into account. Many transport providers are already betting Connection planner to in addition to passenger information to solve additional operational tasks, e.g. around the Use drivers or vehicles to plan or traffic to control or to send the printed timetables to the To generate breakpoints. These connection planners are often proprietary solutions that the transport providers for their respective requirements and operational conditions let develop. For the development and customization of his The traffic provider usually has considerable connection planners Have to go to great effort, for example to coordinate everyone involved and troubleshooting.

If a traffic provider has a traffic information system want to provide that takes into account interference, it is therefore of great advantage if he is free to choose the Connection planner has. In particular, it is advantageous to use one to continue to use the existing connection planner and to be able to supplement the consideration of faults. EFA and any other traffic information system according to the state of the Technology does not meet these requirements because they have each have their own connection planner, which is used before a mission only to the particular circumstances and requirements of one Transport provider must be tailored.  

The invention is based on the following object: a Traffic information system to be provided current disruptions in the determination of connecting journeys automatically taken into account and that the information about the undisturbed traffic network automatically from any Connection planner wins.

The object of the invention is achieved by two methods for automatic determination of connecting trips using two data processing systems, namely one Connection planner and a fault manager. The central one Idea is that these two data processing systems are independent operate from each other. The connection planner "knows" the valid timetable for the undisturbed network, but no current one Disorders. The fault manager "knows" at least that relevant of the currently existing faults, but not the current timetable. Connection planner and fault manager with each other via a third data processing unit coupled, namely a coordination unit. The invention teaches how the traffic information system automatically Connections with information about all faults these connecting trips as well as trouble-free connecting trips certainly.

A crucial feature of the invention is that a Connection planner that the transport provider already has used, can still be used. On Connection planners are typically addressed to the specific Needs and conditions of the transport provider and / or adapted to his customers. The desired properties remain obtained by the method according to the invention, in addition = current faults taken into account. Because that Traffic information system is basically only those  Connections that the currently available Connection planner has calculated. The invention thus solves the underlying task.

The fault manager preferably "knows" the topology of the Transport network, so it has read access to a list of all Stops and routes of the transport network. If one Connection journey is affected by a malfunction, so the fault manager determines a path from the specified one Starting point to the specified destination, which is trouble-free, and the connection planner finds a connection journey via this Path.

Further advantageous embodiments are in the subclaims reproduced.

The invention is based on exemplary embodiments in the Drawing figures explained in more detail. Show it:

Fig. 1 configuration of a traffic information system according to the invention;

Fig. 2 advantageous embodiment of the invention;

Fig. 3 method for determining trouble-free connecting trips;

Fig. 4 embodiment of the method for determining trouble-free connecting trips.

Fig. 1 describes a traffic information system that operates according to the inventive method. The traffic information system 2 consists of a connection planner 4 , a fault manager 6 and a coordination unit 8 . The connection planner 4 and the fault manager 6 are information processing systems which operate independently of one another. The coordination unit 8 is connected to both the connection planner 4 and the fault manager 6 .

The coordination unit 8 receives the desired starting point 5 and the destination point Z, that is to say those nodes of the transport network at which the journey is to begin or end. S and Z can be stops of the transport network or instead, for example, geographical points that are specified by length and latitude, or postal information (street, house number, postcode, city). In addition, the coordination unit 8 receives all the necessary information about time restrictions which the user places on the journey, for example the earliest or latest time of the beginning or end of the journey or the maximum duration of the journey. The following sequence is carried out at least once: The coordination unit 8 activates the connection planner 4 . This determines a connecting journey Vf from S to Z that meets the specified time restrictions and can be implemented with the valid timetable. The individual connections of this connection journey Vf are then checked whether they have been checked by a current fault or not. For this test, the malfunction manager 6 evaluates information about current malfunctions in the traffic network. The result of the check of Vf is transmitted to the coordination unit 8 , which outputs the result. Depending on the design of the traffic information system 2 , this sequence can be run through once or several times, which is described in more detail in the advantageous refinements.

So that the fault manager 6 can check individual connections as described above, the faults must be recorded in a structured form. In particular, for each fault, the affected stop or the affected route between two stops and the fault period must be made known to the fault manager 6 . The prior art knows methods and devices for how these requirements can be met, for example pre-formulated input masks for a human processor or at least partially automated analysis of a description of the fault formulated as free text. Furthermore, when designing a traffic information system according to the invention, one must define a data format in which the connection planner 4 transmits a connection to the fault manager 6 . The prior art knows methods and devices to record faults in a structured form and to define a common data format. If these two requirements are met, the invention provides a traffic information system that takes current faults into account and uses any connection planner 4 . The invention therefore achieves the above-mentioned object.

A traffic information system 2 according to claim 1 works in such a way that the connection planner 4 determines one or more connection trips and the fault manager 6 checks for each of these connection trips whether the trip is impaired by faults. The connection trips together with the identified faults are output. This traffic information system gives the user the opportunity to decide for himself whether and if so how to take the disturbance into account.

Connection journeys and faults are preferably determined in such a way that the connection planner 4 first determines a connection journey Vf from S to Z and then the fault manager 6 checks whether Vf is impaired by a current fault. This sequence is run once or several times, depending on the design. The coordination unit 8 outputs the connection trips determined in this way and the disturbances affecting them.

Some disturbances consist of the fact that a means of transport is a Route of the transport network only slower and therefore only with one Can be delayed compared to the timetable or one Breakpoint can only happen with a delay. On Example: On a route between 2 stops Track construction work carried out, which is why the route with at most 50 km / h can be driven and the travel time increases by a maximum 5 minutes longer. An advantageous embodiment (claim 3) The invention provides a method for this situation Provides traffic information.

It is assumed that the fault manager 6 has read access to information about the maximum duration of the delay for each fault that is a delay. The connection planner 4 first determines, as usual, a connection journey Vf from the starting point S to the destination point Z, without taking disruptions into account. The fault manager 6 determines whether current faults will lead to delays on this connection journey Vf. Let H be the stop at which a delay affects for the first time in the course of Vf. The fault manager 6 adds the maximum duration of the delay to the departure time in H, which the connection planner 4 has determined for Vf. The coordination unit 8 then activates the connection planner 4 again, this time with the aim of determining a connection journey from H to Z with the departure time, which is corrected by the delay, ie postponed in time. Following the same procedure, further delays at the following stops will be taken into account. Overall, the traffic information system delivers a connecting journey that takes into account all delays.

The embodiment of the invention just presented can be used in particular if the earliest departure time from S is specified for the traffic information system. Another embodiment (claim 4) comes into play when the latest arrival time has been specified in Z. After the traffic planner 4 has determined a connection trip from S to Z without taking disruptions into account, the fault manager 6 determines the stopping point H at which a delay will occur for the last time. The connection planner 4 is activated with the aim of determining a connection journey from S to H with the latest arrival time, which is corrected by the delay, ie postponed in time. Further delays at previous stops are taken into account according to the same scheme.

Sometimes it is sufficient for the traffic information system 2 to provide one or more connection journeys between the starting point S and the destination point Z and disturbances on these connection journeys. Most of the time, however, the traffic information system will be required to determine and output an undisturbed connection journey. Claim 5 teaches a method to achieve this important part of the object on which the invention is based.

Just as in the method according to claim 1, connection planner 4 and fault manager 6 work independently of one another, and fault manager 6 "knows" the current faults, connection planner 4 knows the schedule. The coordination unit 8 accepts the request. The connection planner 4 determines, once or repeatedly, connection journeys from the specified starting point S to the specified destination point Z. These connection journeys are checked for faults, for which purpose the information to which the fault manager 6 has read access is evaluated. The coordination unit 8 outputs the connection journeys recognized as being trouble-free or reports that no currently trouble-free connection journeys from S to Z could be determined. The traffic information system has thus completed its work.

Preferably (claim 6) the sequence is run through at least once that the connection planner determines 4 connection journeys from the specified starting point S to the specified destination point Z and the fault manager 6 then checks these connection journeys for faults. If one of these connection journeys is impaired by a fault, the connection planner 4 determines further connection journeys from S to Z which deviate from the connection journeys determined so far. This sequence is repeated as often until interference-free connection journeys have been determined or until it is certain that there is no interference-free connection journey from S to Z.

The search for trouble-free connecting journeys can take a long time. In order to limit the computing time and thus to ensure that a device operating according to the method according to the invention reaches a result in good time, it is advisable to define a termination criterion in advance. If this criterion is met, the traffic information system stops its work, even if no trouble-free connection journey was found (claim 7). Two sensible termination criteria are:

• - Abort if the fault manager has checked 6 N connection trips, N is a predetermined natural number
• - Abort when the computing time has reached T seconds.

An operation of the fault manager 6 is described by an advantageous embodiment ( Fig. 2). The fault manager 6 has read access to information about all stops of the traffic network and all existing routes between them. This information can easily be obtained from that used by the traffic planner. They are expediently kept in a database, namely the topology database 10 . Other storage media and types, such as files, can also be used. In contrast, the fault manager 6 does not "know" a timetable for the traffic network. The fault manager 6 keeps a constant record of all faults that have occurred. The information about faults that have occurred are recorded - for example by manual entry or by an operating control system automatically transmitting the information - and then also stored in a database, the fault database 12 , to which the fault manager 6 has read access. If necessary, the information in this database is updated.

Different configurations determine which information about a fault is stored. They include:

• - Specification of all stops and routes that are marked by a Disturbance
• - Indication of all route sections caused by a fault be affected
• - the impaired direction of travel if the fault changes affects only one direction of travel on a route
• - the period during which a fault occurs
• - the type of disorder, e.g. B. Delay.

An advantageous embodiment of this invention (claim 8) specifies an efficient method for determining trouble-free connecting journeys. The designations of Fig. 2 are used.

If a connection journey Vf is impaired by a fault, the fault manager 6 determines a fault-free path between S and Z. This path consists of stops and routes in the traffic network which are not impaired by current faults. The prior art knows various methods under the keyword routing methods, which automatically find a path from S to Z, if there is such a path at all. The traffic network is interpreted as a directed graph. The nodes stand for the stops and the edges for the routes of the transport network. The fault manager 6 applies a routing method to the traffic network without the disturbed routes and stops. This path does not depend on the timetable, but only on the topology of the transport network.

If the malfunction manager 6 does not find a trouble-free path from S to Z, the traffic information system 2 outputs the result that no trouble-free travel from S to Z is currently possible. If it has determined a trouble-free path Pf from S to Z, then the traffic information system 2 then determines an undisturbed connection journey according to the invention. The path Pf itself is not a connection path, since it defines stopping points and routes, but no individual connections, that is to say no departure times and no lines - the fault manager 6 only "knows" the traffic network, but not the timetable. Therefore, the connection planner 4 is activated again. In order to avoid that the traffic planner 4 selects the faulty connection journey Vf again, the fault manager 6 selects a stop H which is on the path Pf but is not touched by the connection journey Vf. The traffic planner 4 is activated with the query to deliver a connection journey from S to Z via H, that is to say a partial connection journey Vf1 from S to H and a further partial connection journey Vf2 from H to Z. As a result, the traffic planner 4 is "forced" to determine a different connection journey from S to Z as Vf. The fault manager 6 then checks whether the two partial connection journeys Vf1 and Vf2 are fault-free. If this is the case, the result is the connection journey, which consists of Vf1 and then Vf2. If Vf1 is disturbed, the above procedure is repeated with Vf1 instead of Vf. The fault manager 6 thus determines a fault-free path Pf1 from S to H and then a breakpoint H1 on Pf1. The connection planner 4 is then activated with the aim of determining a connection journey from S to H via H1. These operations are repeated until a trouble-free connection journey from S to Z is found or until the traffic information system 2 has determined that there is no such connection journey.

In the advantageous embodiment according to claim 8 of the invention, the routing method was only required to guarantee that it will find a path from S to Z if such a path exists at all. Another embodiment (claim 9) specifies that a routing method is used which finds an optimal path from S to Z. The criterion for which the path should be optimal must be specified. Useful criteria are the geographical length of the path, the travel time or the costs that the passenger incurs when traveling along this path. The traffic network is understood as a directed graph, the edges of which are weighted. A path is optimal if the sum of the weights of all the edges that make up the path is minimal. The prior art knows several routing methods that provide paths that are optimal with respect to given weights. In order to use such a routing method, the fault manager 6 must know a "weighting" for each route. A weighting is therefore stored in the topology database 10 for each route and each stopping point of the traffic network N. The weighting is determined by the given optimality criterion. If you choose the first of the above three criteria, the weighting depends on the travel time of each route and the waiting time at each stop, the second criterion on the geographical length of each route and the transfer route at each stop. Note: Although using an optimal path does not guarantee that the method determines the fastest, shortest or cheapest connection journey, good connection journeys can be expected in practice.

An embodiment of the method (claim 10) with which interference-free connection journey is found, takes into account the The fact that some faults only last for a certain period of time occur and that the traffic information system has read access to Has information about that period. An example: the Transport provider designed his vehicle fleet and his Organization designed so that it always takes a maximum of 30 minutes until a failed vehicle is cleared away and a replacement vehicle is provided. If there is a current malfunction, that a vehicle has "stopped" on a certain route is, this disturbance will last a maximum of 30 minutes. Be Pf a path in the traffic network N, which leads over a route S, the is disturbed in a certain period of time. A connecting trip on this path there is only interference in S if the Connection journey used during this period S, outside of this Period S is trouble-free.

According to the embodiment according to claim 10, the minimum and maximum travel time of each route as well as the minimum and maximum waiting time in each stop, which contains the transfer time in this stop, are stored in the topology database 10 . The travel and transfer time can vary depending on the time of day or season. The information about travel and transfer times can often be obtained automatically from information that transport providers have in their computer systems - for example from journey time profiles for lines.

The fault manager 6 searches for a fault-free path from S to Z, so that a fault-free connection journey from S to Z determined by the connection planner 4 and impaired by a current fault is subsequently determined. The procedure by which this path Pf is searched from S to Z is structured as follows: At t0, the departure time from Vf, that is, the time at which the connecting journey begins in S. In a routing method, a partial path from S to an intermediate stop is repeatedly determined and a continuation, ie the next route and the next stop, is sought. At Pf1, a path that begins in S and that is to be checked for faults. The fault manager 6 determines for each route and each stop of Pf1 a period of time during which a means of transport that starts in S at time t0 travels this route or approaches this stop. This calculation is carried out by adding waiting times for the lines and stops from Pf1 to t0. If the time period determined in this way for a route does not overlap with the time period for a disruption on this route, the route is trouble-free, otherwise disrupted, and the same applies to every stop. Pf1 is marked as trouble-free if all routes and stops of Pf1 have been recognized as trouble-free according to the procedure described above.

Expediently (claim 11) the fault manager objects step-by-step routing process to create a path of  S to be determined by Z. Such procedures in connection with a Predefined optimality criteria are often used in the literature referred to as a process that is a "best-first strategy" - too called "greedy strategy" - apply. Typically is one Criterion specified to evaluate paths - in the case of a Transport network this can be the minimum or maximum travel time, the longitude or the cost of travel. At a step-by-step routing procedures are used in every step alternative partial paths Pf_1,. . ., Pf_n determined in S begin and in intermediate stops H_1,. . ., H_n ends that are different in pairs. An intermediate breakpoint H_i becomes selected; the different procedures differ essentially in terms of which criterion the Selection is made. Most routing procedures do the search ends, if H_i is Z, then Pf_i is a path from S to Z. Otherwise let N_1,. . ., N_m all or some selected neighboring breakpoints of H_i, i.e. breakpoints that are directly connected to H_i by a route. The path Pf_i is extended so that the m paths Pf_N_1,. . ., Pf_N_M arise that begin in S, go over H_i and in the Intermediate stops N_1,. . ., N_m ends. It is checked whether also under the "old" paths Pf_1,. . ., Pf_i-1, Pf_i + 1, . . ., Pf_n are paths that are in one of the "new" intermediate Breakpoints N_1,. . ., N_m ends. If so, either will "old" path Pf_j or a "new" path Pf_N_j deleted - each after which path is rated as worse. The Execution of the routing procedure is the next step with the remaining paths from the set Pf_1,. . ., Pf_i-1, Pf_N_1,. . ., Pf_N_M, Pf_i + 1,. . ., Pf_n continued.

Fig. 4 illustrates this process. Before the illustrated step of the routing method, the three paths Pf_1 (from S to H_1), Pf_2 (from S to H_2) and Pf_3 (from S via X to H_3) were determined. The routing procedure selects H_3. N_1 and N_2 are the two neighbors of H_3. The next step is the search with the four paths Pf_1. (from S to H_1), Pf_2 (from S to H_2), Pf_N-1 (from S via X and H_3 to N_1) and Pf_N-2 (from S via X and H_3 to N_2).

An example of such a method is the A * algorithm. The point H_i is selected for which the evaluation of the Paths from S to H_i plus the estimated path rating from H_i to Z is best. Accordingly, one is required Estimate for the evaluation of a path from H_i to Z. Eine An evaluation is proportional to the length of the design Distance from H_i to Z, z. B. Air line multiplied by Average speed of transportation. Is the Estimated path rating always less than that the A * algorithm actually provides an optimal path from S to Z. A special case of the A * algorithm is Dijkstra 's algorithm where the estimated rating of the Path from H_i to Z is always 0.

The fault manager 6 expediently uses a step-by-step routing method as follows: at H_i the selected end point. The fault manager 6 determines the earliest and the latest departure time of a means of transport in HJ that starts in S at time t0. T0 is the point in time at which the faulty connection journey Vf begins in S. An adjacent breakpoint I_j is only taken into account if, as described above, there is no disturbance in the period to be taken into account neither on the route from H_i to I_j nor in I_j.

An embodiment of the invention (claim 13) takes into account the possibility that faults occur in certain periods of time and are then eliminated. These periods of malfunctions are stored in the malfunction database 12 .

Such a period consists of either only a start or an end time or a start and end time. Whenever the fault manager 6 checks a connection trip determined by the connection planner 4 for faults, he compares the departure and arrival times of the individual connections with the periods of the faults.

Various advantageous refinements of the invention define which information about faults is stored in the fault database 12 and how the fault manager 6 processes this information in order to check whether a connection trip is impaired by a fault.

An embodiment (claim 14) specifies that it is stored which stops and routes in the transport network are impaired by a fault. The entry can of course consist of only a single stop or a single route. In addition, a description of the topology of the traffic network N is stored in the topology database 10 , that is a list of all the stops and routes of N. The fault manager 6 "receives" a connection trip from the connection planner 4 for testing. This consists of individual connections. Information about a single connection includes its start and end times. The fault manager 6 determines which routes and breakpoints it uses for each individual connection, for this purpose it uses the information in the topology database 10 . Then he determines which current disruptions affect these routes and stops. In this way, he determines which faults exist on the individual connection. Overall, the fault manager 6 determines in this way which faults are present on the connection journey that the connection planner 4 has transmitted.

In order for the malfunction manager 6 to work quickly, the malfunctions should be stored in the malfunction database 12 in a manner that enables quick access. In a traffic information system 2 according to the invention, it is frequently necessary to check whether an individual connection is impaired by a current fault or not. For this purpose, the routes and stops via which the individual connection leads must be checked for faults. According to the advantageous embodiment (claims 15), the disturbances are stored in the disturbance database 12 sorted according to the stop and route concerned. This database is not updated only when a connection journey is to be determined, but whenever a new fault is reported or an "old" fault has been eliminated.

Often the reason for a malfunction is that a certain road or a single section of track cannot be used. The disruption then affects all routes in the transport network that run over this road or section of track. It is therefore advantageous to assign the affected route sections directly to these disruptions and to have the traffic information system 2 determine automatically what effects this disruption has on the traffic network. According to the embodiment according to claim 16, the fault database 12 stores which route sections are impaired by a fault. The fault manager 6 "knows" by means of read access to the topology database 10 , via which route sections an individual connection leads, and determines - analogously to the embodiment according to claim 15 - which faults impair a connection journey. The faults are expediently stored in the fault database 12 sorted according to route sections (claim 17).

"Agent-oriented programming" is a procedure around distributed, concurrent processes To let information processing units run. The individual information processing units operate autonomous, d. H. without central control. You will software Called agents. The advantages of doing this are: Different, heterogeneous information processing units can operate together and jointly desired technical Get results without looking at the individual units design or synchronize homogeneously, which is time consuming and is expensive. The individual units can be independent be expanded, improved and modified from one another without that other units have to be changed. Another Advantage: software agents can be running while Switch operation on and off without interrupting operation have to. This property enables fault tolerant systems to construct.

Implementing a traffic information system according to the invention with the aid of software agents is particularly expedient if the traffic information system is to integrate heterogeneous databases and other data processing systems. Reasons for this heterogeneity can be: A traffic information system should provide connecting journeys that involve several transport providers (e.g. Deutsche Bahn, Berlin S-Bahn and Berliner Verkehrsbetriebe). The IT environment of a transport provider has grown "historically" and is therefore heterogeneous. The advantageous embodiment according to claim 18 teaches to construct the connection planner 4 , the fault manager 6 and the coordination unit 8 using software agents.

So that the malfunction manager 6 can manage the information about current malfunctions and compare them with connecting journeys, they must be made available to him in a computer-available form. In principle, it is possible according to the prior art that a text analysis program automatically obtains this information from free text. However, fewer errors and less time are required if the information about faults is available in a structured form from the outset. In many cases, a human operator will have to provide this information to the fault manager, since the computer is not yet available. A data input device, as is known in the prior art, is expediently made available to him. This device has its own input field for each information required, in particular for the routes and stopping points affected by the fault, the affected route sections, the period of the fault and the type of fault. The human processor enters free text in the input field or selects the disturbed breakpoints from a specified list of breakpoints.

The invention cannot only be used to: Build traffic information system that is one for passengers Traffic network N determines and outputs connecting trips. she can also be used in a transport network or also in a production facility, timetable-based means of transport (Use claim 20) or transport systems (Use claim 21) to control. The results of the Traffic information system will not become a passenger displayed, but used to other facilities too Taxes. Two examples are described below.

At Frankfurt / Main Airport and other airports Driverless means of transport operate as "people mover" be designated. These modes of transport control various Stops at the airport site. From operational  The "people movers" may only give reasons for certain reasons Use certain sections of the transport network for periods of time. These restrictions are in the form of the control device provided a schedule that allowed the Describes trips. The trips mentioned in the timetable are only the possible, not the actual trips. With everyone The transport request determines the control, which after the works according to the invention, a connecting journey, thus a permissible journey in the traffic network.

Driverless transport systems are vehicles that are typically Transport workpieces in a hall, e.g. B. Bodies in an assembly plant for motor vehicles. They drive - according to the Specifications of a control device and influenced by the Measurement results from sensors - automatically one after the other various processing stations and warehouses to make workpieces pick up or drop off. In flexible manufacturing plants If the workpieces are not processed according to a fixed cycle, but depending on the demands for the produced ones Workpieces that can fluctuate in time (keywords "just in time "and" lean production ").

A connecting trip determines which stops, i.e. H. which stations and warehouses a driverless transport system too what time is coming. The routes between the Stops are the routes that the transport systems use allowed to. Examples of disturbances at stops and routes are: A processing station has failed, a workpiece got stuck, a path is blocked, a driverless one Transport vehicle has failed. The invention provides Process with which driverless transport systems also then can be controlled efficiently when faults occur are.  

Overview of terms used Breakpoint

General term for stop (for bus, tram and trucks), airport, train station, Landing stage (for ferry) and for Processing station and location (for Production plant)

route

Generic term for the connection between two directly and stops connected without stopping of a transport network. Driving on a route at least one line.

Route section

A route runs over one or more Route sections. A route section is one Street or part of a street (bus, Tram) or a section of track (railway) or a single route (production plant)

Connecting trip

A connecting journey between an initial and an end stop, both of which are predetermined, consists of one or more individual connections.

Single connection

A single connection consists of an initial and an end stop, a line, and that Departure time at the start stop and the Arrival time at the end stop. A Single connection can be made according to the valid timetable when there is no malfunction, without changing trains be driven on.

Claims (22)

1. A method for the automatic determination of connection journeys in a schedule-linked traffic network N, wherein at least the starting point S and destination point Z of the desired connection journey is specified for the connection journey, characterized in that

• a first information processing unit (connection planner <( 4 ) <) is used, which determines at least one connection journey from S to Z, taking into account a timetable,
• a second information processing unit (fault manager <( 6 ) <) is used, which evaluates information about faults that have occurred in the traffic network N,
• - The connection planner ( 4 ) and the fault manager ( 6 ) operate independently and
• - for at least one connection journey Vf from S to Z, which the connection planner ( 4 ) has previously determined, it is checked whether this connection journey is impaired by a fault,
• - and at least the connection journey Vf is output with details of all individual connections and the faults occurring on these individual connections.

2. The method according to claim 1, characterized in that

• - the following operations a) and b) are carried out at least once:
  • a) the connection planner ( 4 ) determines a connection journey Vf from S to Z,
  • b) the fault manager ( 6 ) checks whether the connection journey Vf is impaired by a current fault.

3. The method according to claim 1 or 2, characterized in that

• - The malfunction manager ( 6 ) for at least one malfunction in the traffic network N, which is a delay, has read access to information about the maximum duration of the delay
• - And the fault manager ( 6 ) at least one connection journey Vf, which the connection planner ( 4 ) has determined and which is affected by a delay, is replaced by a connection journey in which this delay is taken into account by a later departure time from a stop of the connection journey Vf.

4. The method according to any one of claims 1 to 3, characterized in that

• - The malfunction manager ( 6 ) for at least one malfunction in the traffic network N, which is a delay, has read access to information about the maximum duration of the delay
• - And the fault manager ( 6 ) at least one connection journey Vf, which the connection planner ( 4 ) has determined and which is affected by a delay, is replaced by a connection journey in which this delay is taken into account by an earlier arrival time in a stop of the connection journey Vf.

5. A method for the automatic determination of connection journeys in a schedule-linked traffic network N, wherein at least the starting point S and destination point Z of the desired connection journey are specified for the connection journey, characterized in that

• a first information processing unit (connection planner <( 4 ) <) is used, which determines at least one connection journey taking into account a timetable,
• a second information processing unit (fault manager <( 6 ) <) is used, which evaluates information about faults that have occurred in the traffic network N,
• - The connection planner ( 4 ) and the fault manager ( 6 ) operate independently of one another, and
• - either at least one connection trip is output with details of all individual connections that the fault manager ( 6 ) has marked as fault-free,
• - or the result is output that no trouble-free connection travel from S to Z has been determined.

6. The method according to claim 5, characterized in that

• - the following operations a) and b) are carried out at least once:
  • a) the fault manager ( 6 ) checks whether a connection trip Vf from S to Z, which the connection planner ( 4 ) has previously determined, is impaired by a current fault,
  • b) if Vf is impaired by a current disturbance, the connection planner ( 4 ) determines a connection journey from S to Z that differs from Vf.

7. The method according to claim 5 or 6, characterized in that the coordination unit ( 8 ) aborts the determination of connecting trips when

• - either a connection journey from S to Z was determined that is not affected by a fault,
• - or a previously defined termination criterion is met.

8. The method according to any one of claims 5 to 7, characterized in that

• the fault manager ( 6 ) has read access to a first database (topology database <( 10 ) <), in which at least one list of all breakpoints and all routes between two adjacent breakpoints is stored,
• - The malfunction manager ( 6 ) has read access to a second database (malfunction database <( 12 ) <) in which information about at least one malfunction St of the traffic network N which affects schedule operation is stored, at least the information being stored as to which breakpoints and routes of the traffic network N are impaired by the disturbance St,
• - and at least once the following operations a) to j) are carried out:
  • a) the connection planner ( 4 ) determines a connection journey from S to Z,
  • b) the fault manager ( 6 ) checks whether Vf is impaired by a current fault,
  • c) if Vf is trouble-free, Vf is output as a result and the operations are terminated,
  • d) if Vf is impaired by a current fault, the fault manager ( 6 ) searches for a path from S to Z using the information stored in the topology database ( 10 ) and using a routing method, which is exclusively not caused by faults impaired stops and routes of the transport network N,
  • e) if the fault manager ( 6 ) cannot find such a path, the result is output that there is no connection travel from S to Z which is not impaired by faults, and the operations are terminated,
  • f) if the fault manager ( 6 ) finds a fault-free path from S to Z, the fault manager ( 6 ) selects a breakpoint H on this path,
  • g) the connection planner ( 4 ) determines a connection journey Vf1 from S to H and a further connection journey Vf2 from H to Z,
  • h) the fault manager ( 6 ) checks whether Vf1 or Vf2 are affected by a current fault,
  • i) if both Vf1 and Vf2 are not affected by interference, the operations are terminated
  • j) if Vf1 is affected by a current disturbance, operations d) to i) are performed for Vf1 instead of Vf, and if Vf2 is affected by a current disturbance, operations d) to i) are performed for Vf2 instead of Vf .

9. The method according to claim 8, characterized in

• - That a weighting is stored in the topology database ( 10 ) for each route and every stopping point of the transport network
• - And the malfunction manager ( 6 ) in operation d) uses a routing method that provides a path from S to Z, for which the sum of the weights of all routes and breakpoints contained is minimal.

10. The method according to claim 8 or 9, characterized in

• - that in the topology database ( 10 )
• - The minimum and maximum travel time on this route is saved for each route
• - The minimum and maximum waiting time for each breakpoint is stored in this breakpoint
• - In the malfunction database ( 12 ) for at least one malfunction, information about the period in which this malfunction occurs is stored, and
• when operation d) is carried out at least once when it is checked whether a path Pf1 beginning in S is impaired by a current fault,
• - The minimum and maximum travel times of all routes and the minimum and maximum waiting times of all stops from Pf1 are added to the departure time of Vf
• - And the periods determined in this way are compared with the periods for malfunctions about which information is stored in the malfunction database ( 12 ).

11. The method according to claim 10, characterized in that

• - The fault manager ( 6 ) uses a step-by-step routing procedure and
• when carrying out operation d) at least once when it is checked whether a path Pf1 beginning in S can be extended from a breakpoint H to an adjacent breakpoint I or not,
• - The earliest and the latest departure time in H is compared with the time periods for disruptions that affect the route from H to I and with the time periods for disruptions that affect the stopping point I.

12. The method according to any one of claims 8 to 11, characterized in that

• - The fault manager ( 6 ) applies an A * algorithm when performing operation d).

13. The method according to any one of claims 1 to 12, characterized in that

• - Information is stored in the malfunction database ( 12 ) for at least one malfunction about the period in which this malfunction occurs, and
• - The fault manager ( 6 ) compares the departure and arrival times of the individual connections with the information about periods of faults when checking which individual connections are affected by which faults
• - The fault manager ( 6 ) only identifies an individual connection as being affected by a fault St if the fault St
• - either no information about the period in which this fault occurs is stored
• - or either the departure or arrival time of the individual connection or both times fall within the period of the fault.

14. The method according to any one of claims 1 to 13, characterized in that

• the fault manager ( 6 ) has read access to a first database (topology database <( 10 ) <), in which at least one list of all breakpoints and all routes between two adjacent breakpoints is stored,
• - The malfunction manager ( 6 ) has read access to a second database (malfunction database <( 12 ) <) in which information about at least one malfunction St of the traffic network N which affects schedule operation is stored, at least the information being stored as to which breakpoints and routes of the traffic network N are impaired by the disturbance St,
• - the fault manager ( 6 ) for at least one connection journey Vf, which the connection planner ( 4 ) has previously determined,
• - checks which stops and routes belong to the individual connections of this connecting journey Vf,
• - and then determines which current disturbances are affecting these stops and routes,
• - and then determines which current disturbances affect the connection journey Vf.

15. The method according to claim 14, characterized in that

• - The faults are stored in the fault database ( 12 )
• - That for each stop of the transport network N information about all disturbances that affect this breakpoint are stored, and
• - that information about all disruptions affecting this route is stored for each route between two stops,
• - And whenever an additional fault has occurred or a previously existing fault has ended, the fault database ( 12 ) is updated.

16. The method according to any one of claims 1 to 15, characterized in that

• - the fault manager ( 6 ) has read access to a first database (topology database <( 10 ) <),
• - In the at least one list of all stops of the transport network N and all routes between two neighboring stops
• a list of all the route sections that make up this route is stored for each route,
• - The malfunction manager ( 6 ) has read access to a second database (malfunction database <( 12 ) <) in which information about at least one malfunction St of the traffic network N which affects schedule operation is stored, at least the information being stored as to which route sections a route of the traffic network N is impaired by the disturbance St,
• - the fault manager ( 6 ) for at least one connection journey Vf, which the connection planner ( 4 ) has previously determined,
• - checks which route sections belong to the individual connections of this connecting journey Vf,
• - and then determines which current disturbances are affecting these route sections,
• - and then determines which current disturbances affect the connection journey Vf.

17. The method according to claim 16, characterized in that

• - The faults are stored in the fault database ( 12 )
• - That for each stop of the transport network N information about all disturbances that affect this breakpoint are stored, and
• that information about all disturbances affecting this route section is stored for each route section of a route between two stops,
• - And whenever an additional fault has occurred or a previously existing fault has ended, the fault database ( 12 ) is updated.

18. The method according to any one of claims 1 to 17, characterized in that

• - the connection planner 4
• - and the fault manager 6
• - be constructed using software agents.

19. The method according to any one of claims 1 to 18, characterized in that

• a fourth information processing device (data collector) is used,
• - With the aid of the data logger, at least one of the following information can be entered for each current fault in a separate field for data entry:
• - the routes affected by the disruption
• - the breakpoints affected by the disturbance
• - The route sections that are affected by the fault
• - the period in which the fault occurs
• - the type of disturbance.

20. Device for carrying out a method according to one of the Claims 1 to 19. 21. Use of a method according to one of claims 1 to 19 for controlling timetable-based means of transport. 22. Use of a method according to any one of claims 1 to 19 for controlling timetable-based transport systems in a production facility.