Computer-implemented method for predicting evolution of traffic simulations

Abstract

To provide a computer-implemented method for predicting the evolution of traffic simulations.SOLUTION: A method includes: accepting input of traffic data, including data acquired from sensors within a geographic region; running traffic simulations using the traffic data for the geographic region; selecting an intervention from a plurality of potential interventions in response to a decision to model the effect of a change to the traffic simulations; modifying parameters of the traffic simulations using the intervention and obtaining the modified simulation parameters; and implementing the modified traffic simulations using the modified simulation parameters and configured to reflect the effect of the intervention on the traffic.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to traffic simulation. More specifically, the present invention relates to a method for predicting the evolution of traffic simulation, and related data processing apparatuses, computer programs, and computer-readable storage media.

Background Art

[0002] There is a tendency for the real world to be represented as a digital world. Concepts such as smart cities, digital twins (DT), and the metaverse have received particular attention, at least due to new and improved technologies that enable progress within these concepts, such as the Internet of Things (IoT), the 5th generation technology standard for telecommunications (5G), artificial intelligence (AI), and quantum computing.

[0003] Regarding improvements in transportation, smart transportation systems are being used to provide a wide range of services to various users on a large scale and in real time (or fast enough for effective actions to be taken). The system desirably provides a wide range of services created by complex processing of mainly real-time detected data obtained from sources (such as roads and vehicles) within the transportation network (or providing services to the transportation network). Services are often required "in real time" so that immediate actions can be taken by the users of the services. There are various data sources that generate different types of data at different rates. Services also use data in a wide variety of different ways and are diverse and complex.

[0004] As an example, a transportation agency in a large city may create a simulation in the form of its public transportation infrastructure system DT that provides a real-time virtual model of the traffic flow across the city. The virtual model can be driven by data generated in real time from a wide range of sources, including vehicle movements from in-vehicle GPS, roadside sensors, CCTV analytics, routing requests, live dispatching of public transportation, etc., the status and operation of infrastructure such as traffic lights, and people's movements from mobile GPS, ticket issuance activities, CCTV, etc.

[0005] System DT (the virtual model of the system) integrates data sources and provides a real-time understanding of the traffic flow state. Synchronized services may be added to DT, for example, using anomaly detection methods to determine events (such as accidents, congestion, etc.). However, there is also a need to manage such events, for example, to divert traffic around an accident site, deploy emergency services, and / or change the routes of public transportation.

[0006] The requirements of a smart transportation system are to predict the state of the system and enable services. However, traditional simulations used for traffic prediction are typically fixed with respect to goals and behaviors. Traditional simulations are not sufficiently predictive or adaptable enough to adapt to potential scenarios not modeled by the original simulation. Rather, such techniques simply emulate the modeled scenarios. To simulate new behaviors, an entirely new simulation model should be created, which is computationally difficult and time-consuming to prepare and start. This is not optimal for a real-time environment.

[0007] Among the state-of-the-art technologies developed to address this issue is deep learning (DL) simulation modeling, which utilizes adaptive machine learning models. However, these rely on well-constructed training datasets that are cumbersome to prepare. The intent of DL simulation (the effect of new behaviors on the simulated system) is actually fixed to the training dataset, and the intent does not evolve if the behavior is outside the range. Also, training such specific models is computationally expensive, and simulation models have to handle large amounts of diverse data.

[0008] Accordingly, the inventor has come to the recognition that there is a need for a technology to improve the realistic simulation of scenarios in the digital representation of the real world. SUMMARY OF THE INVENTION

[0009] The present invention is defined in the independent claims and should be referred to therefrom. Further features are described in the dependent claims.

[0010] According to an aspect of the present invention, there is provided a computer-implemented method for predicting the evolution of traffic simulation, the method comprising receiving an input of traffic data including data obtained from sensors within a geographical area, performing a traffic simulation using the traffic data of the geographical area, selecting an intervention from a plurality of potential interventions in response to a decision to model the effect of a change to the traffic simulation, changing the parameters of the traffic simulation using the intervention and obtaining the changed simulation parameters, and performing a changed traffic simulation configured to reflect the effect of the intervention on traffic using the changed simulation parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Embodiments for Carrying Out the Invention

[0012] As a mere example, refer to the accompanying drawings.

[0013] In actual use cases of known simulation techniques for tracking traffic simulation scenarios, sensor data is supplied to the simulation and a behavior model (e.g., a conventional ABMS) can be reproduced. According to such an approach, it is difficult to reproduce the behavior of future scenarios that are not modeled or not necessarily notified to the simulation system. For example, a sudden increase in pedestrian traffic in an area affected by a concert or other event may not be easily considered by known simulation techniques. Since known simulation systems do not have a means for detecting abnormal future scenarios and are not adaptable enough to adopt new simulation behaviors without generating an entire new model, new future scenarios are not notified and processed by known simulation systems.

[0014] According to an aspect of the present invention, there is provided a computer-implemented method for predicting the evolution of traffic simulation. The method includes the step of receiving an input of traffic data. The traffic data may be vehicle data and / or pedestrian data. The traffic data includes data obtained from one or more sensors within a geographic area (the area of the traffic simulation). The traffic data may further include data regarding current or future weather, seasonality, pre-determined events (e.g., concerts or sports events), mobility flows, and correlation patterns of traffic from other geographic areas (adjacent and non-adjacent geographic areas).

[0015] The method includes the step of performing a traffic simulation using the traffic data of a given area. In this way, traffic predictions can be generated and the manner in which traffic patterns within the geographic area evolve can be represented. Any known traffic simulation technique may be used for this purpose.

[0016] The method includes performing a decision process to determine whether to model or simulate the effects of changes or interventions on traffic simulations. For example, it may be determined to model the effects of interventions within events where the prediction of simulated behavior results in abnormal traffic behavior. In response to a positive decision, the method includes selecting, from among a plurality of potential interventions, the intervention to apply to traffic within the simulation.

[0017] The method includes changing the parameters of the traffic simulation in light of the selected intervention. Thus, the method obtains the changed simulation parameters. The changed simulation parameters are configured to cause the simulation to reflect the effect (or effects) of the selected intervention on traffic.

[0018] The method then includes executing the changed traffic simulation. The changed traffic simulation may use the changed simulation parameters and further traffic data obtained from sensors within the geographic area of interest. Thus, the method may generate a changed traffic prediction that reflects the effect of the selected intervention on traffic.

[0019] Vehicle traffic behavior models and human traffic behavior models play important roles in the field of DT and will be useful in future, for example, for metaverse solutions. The technology of the present application provides a realistic future simulation that is predictive and adaptive to consider abnormal or unexpected future scenarios.

[0020] The technology of the present application can adapt and evolve the intentions in current simulation systems, thereby providing a system that can provide more realistic behavior. The technology provides predictive and adaptive context simulations for abnormal scenarios and predicted scenarios in DT, enabling smart decision-making to anticipate potential new behaviors of the simulation based on new contexts in DT.

[0021] The method of the present application may be applied to an agent-based modeling system (ABMS), and the simulation of the behaviors and interactions of autonomous agents (individuals, or collective entities such as organizations or groups) enables the understanding of the behavior of traffic systems. Broadly, the agents of the ABMS can have their own pre-programmed goals, and the system as a whole can have pre-programmed goals. Agents can be interconnected with other agents, and each can be implemented as a software module or a bot. In a traffic scenario, one agent can represent a traffic intersection or an individual vehicle. In the ABMS used for simulation, the intended modeled behavior does not evolve over time. As a result, whenever the behavior needs to be incorporated into the simulation, a new ABMS model needs to be created, which is not efficient.

[0022] The method provides an accurate and realistic simulation and reproduces future behavior based on potential events (interventions) that affect the real-world behavior of the underlying system. The method enables the provision and integration of smart decision-making in the current scenario and allows the application of potential countermeasures to minimize the potential impact on future scenarios. In fact, the method enables a digital rehearsal with interventions.

[0023] The technology of the present application provides predictability by enabling the recognition of future scenarios that may affect the current reproduction of future behavior. The technology also provides adaptability and enables rapid adaptation simulations for potentially abnormal future scenarios and predicted future scenarios according to user requirements.

[0024] A method for predicting the evolution of traffic simulations enhances the realistic representation in simulated scenarios of traffic simulations that include human mobility. Thus, the method is applicable to any field where traffic simulations are required and to any other scenario where people's mobility flows may be reproduced.

[0025] Optionally, the method may include the step of outputting instructions (instructions based on simulations and interventions) applicable to a geographical area of the real world. For example, following a simulation of a given intervention, if the effect of the intervention is desirable (e.g., if traffic congestion is significantly reduced), the computing means implementing the method may output instructions for implementing the given intervention.

[0026] Optionally, the determination to model the effect of a change to a traffic simulation may include the step of identifying future events in the traffic simulation. Future events may be identified based on sensor data and / or based on extrapolation of the traffic simulation from its current state. The method may then include the step of predicting the impact of future events on the traffic simulation. If the impact exceeds a predetermined threshold, the method may include the step of determining to model the effect of the change to the traffic simulation.

[0027] As a simple example, sensor data may show that while the number of cars merging onto a road within a geographic area is steadily increasing, the number of cars present on the road remains constant, and prediction or extrapolation indicates that the number of cars on a given road will increase. If the number of cars exceeds a predetermined number considered safe to drive on the road, it may be determined to model the effect of some change or intervention to the simulation of the scenario. As another example, sensor data may be used to monitor pedestrian activity at and around a concert or sports event within a geographic area. Traditional traffic simulations with fixed parameters cannot be used to capture the impact on surrounding roads and pedestrian walkways. In contrast, prediction or extrapolation can predict that the geographic area may become very congested, and the computing system implementing it can then determine that the simulation needs to be "corrected" by an intervention.

[0028] In practice, in that case, the method may include the execution of three simulations (or aspects thereof): a simulation of the "normal" behavior using the initial conditions, a simulation (or prediction) of the changed behavior due to the predicted changes in the simulated initial conditions, and a simulation of the countermeasures of the intervention simulated to confirm the effect of the selected intervention on a specific target (e.g., reduction of emissions, alleviation of traffic congestion, optimization of road capacity, etc.).

[0029] Optionally, selecting an intervention from among a plurality of potential interventions includes determining the scope of the traffic simulation to which the intervention is to be applied. The scope may be, for example, a sub-region of the geographical area of interest, a period or timescale during which the intervention may be applied or be effective, and / or the traffic modalities directly affected by the intervention (although other modalities may be indirectly affected). The method then includes calculating a plurality of scopes, each scope corresponding to one of the potential interventions for a given target. Each scope represents the extent to which the corresponding potential intervention is expected to act on or affect some target variable within the traffic simulation. The target may be, for example, a reduction in emissions within a geographical area, a reduction in traffic congestion, a reduction in commuting time, cost savings, or optimization of road capacity. The method then includes selecting the top-scoring potential intervention as the selected intervention to be applied to the simulation. Optionally, a plurality of interventions, for example, the top 3 scoring potential interventions, may be selected. In this case, the modified traffic simulation may be configured to reflect the effects on traffic caused by the plurality of selected interventions.

[0030] Optionally, calculating a plurality of scores corresponding to a plurality of potential interventions is performed by an optimization process. The optimization process is configured to solve an optimization problem for each given potential intervention while minimizing (or maximizing) an objective function to find an optimal solution (or a solution as close to optimal as possible given time and processing capabilities constraints). The optimization process may be configured to form part of the optimization problem to find the optimal actions to be taken in the real world.

[0031] A solution to a potential intervention can be regarded as the score of that potential intervention and provides a measure of the expected impact of the potential intervention for a given set target (e.g., reduction of emissions, alleviation of traffic congestion, optimization of road capacity, etc.). The objective function can represent, for example, the level of traffic congestion, the monetary value associated with traffic flow (e.g., related to the fuel costs of individual drivers), the level of pollutant emissions (e.g., from all vehicles within a geographical area), or the time value associated with traffic flow (e.g., the average commuting time of all vehicles).

[0032] The objective function may be a multi-objective function where each objective can be optimized simultaneously. That is, there may be multiple set targets for the optimization process. In this case, the score of the intervention may be a weighted average of the scores of the underlying individual targets. For example, when the optimization process is performed with respect to two targets (e.g., reduction of overall CO2 emissions and reduction of the commuting time of vehicle users within a geographical area), the score may be a weighted average of the scores calculated for each target. The weighting parameter is pre-set and can represent the "importance" of the target. For example, if it is desired that each of the two targets is satisfied, a weighting parameter of 0.5 may be applied to each score. If one target is prioritized over the other (e.g., emissions are "more important" than commuting time), a weighting parameter of 0.9 may be applied to one target and 0.1 to the other target.

[0033] Optionally, the intervention may be a constraint or limitation applied to any or all of the following values or categories: the number of vehicles within a geographic area (or sub - area thereof), the transport modalities within a geographic area or sub - area (e.g., only buses and taxis on the road are recognized, private cars are not recognized), the position of vehicles within a geographic area or sub - area (e.g., closing a specific road), the direction of movement of vehicles within a geographic area or sub - area (e.g., implementing a reversal of the contraflow lane), the speed of vehicles within a geographic area (e.g., implementing a minimum or maximum speed limit), the number of accessible (i.e., open) lanes of traffic or roads (e.g., closing a traffic lane within a road), the number of accessible stops or stations (e.g., bus stops, railway stations, subway stations, tram stops), and the behavior of traffic lights (e.g., adjusting the timing of traffic light changes).

[0034] Optionally, the effect of the intervention on traffic may be a change or variation in any or all of the following quantities: the number of vehicles within a geographic area or sub - area, the number of each of the multiple transport modalities within a geographic area or sub - area (e.g., the number of cars and the number of buses), the position of vehicles within a geographic area or sub - area, the direction (or directions) of movement of vehicles within a geographic area or sub - area, the speed of each of the multiple vehicles within a geographic area or sub - area, the average speed of each of the multiple vehicles within a geographic area or sub - area, and the minimum speed and / or maximum speed of each of the multiple vehicles within a geographic area or sub - area. The effect of the intervention may be a change or variation in any or all of the following: the level of congestion within a geographic area or sub - area, the number of traffic jams within a geographic area or sub - area, the level of use of the road network within a geographic area or sub - area, the maximum transport capacity within a geographic area or sub - area, and the level of emissions within a geographic area or sub - area. The effect of the intervention may be a change or variation in any or all of the following quantities: the need for additional routes from a starting point to a destination, the need for a change in the direction of traffic flow within a road, the addition of lanes within some roads, the addition of roads, and the reduction or addition of stops for some transport services (e.g., buses, trains, trams).

[0035] Optionally, the method includes receiving an input of past traffic data of a geographical area (to the computing means implementing it). The method may then include performing a traffic simulation while using both the traffic data and some past traffic data. For example, the simulation may use the history of vehicles or pedestrians in a given geographical area together with live data.

[0036] Optionally, the traffic data may have data obtained from the DT of the geographical area. That is, any sensor may be configured to output sensor data to the DT. DT in this context is a digital model of a traffic system that functions as a digital version that is substantially indistinguishable from the traffic system for purposes such as simulation, integration, testing, monitoring, maintenance, etc. of an actual system.

[0037] Optionally, the sensor may be any or all of the following: a GPS sensor or antenna, a cellular network sensor or antenna, a loop sensor or loop detector capable of recording the number of passing vehicles, a camera sensor, and a vehicle parking sensor. The sensor may be an on-board sensor configured to send the position of a vehicle or pedestrian, for example, a GPS sensor or a cellular (telecommunication) sensor provided in or on a vehicle or pedestrian. Triangulation of antenna signals may be used to determine traffic position. Along with GPS data and cellular data, any other floating car data may be appropriately used. The camera sensor may be used to remotely accumulate image data and provide position information of vehicles or pedestrians using image processing techniques such as object detection. The vehicle parking sensor (mounted on the vehicle or located away from the vehicle) may be used to indicate the parking position of the vehicle and provide, for example, the occupancy level of a parking lot.

[0038] Optionally, the step of changing the parameters of the traffic simulation and running the changed traffic simulation may be executed in parallel with the traffic simulation. In this way, there is no need to stop the ongoing simulation. Rather, the traffic simulation continues without being hindered, and the prediction function and adaptation function are provided. The traffic simulation (both the original simulation and the subsequent changed simulation) can be executed at least in real time, that is, the simulation clock moves at least as fast as the actual clock. There may be advantages when the traffic simulation is executed faster than real time. In that case, multiple potential interventions can be investigated and the results can be provided to the user within a time frame that can be normally implemented. In one implementation, the traffic simulation can be executed at 50 times the speed of real time.

[0039] Embodiments of other aspects include a data processing device having a memory and a processor. The memory stores computer-executable instructions for executing a computer-implemented method for predicting the evolution of a traffic simulation. The processor is configured to execute the instructions.

[0040] Embodiments of other aspects include a computer program having instructions that, when executed by a computer, cause the computer to execute a computer-implemented method for predicting the evolution of a traffic simulation.

[0041] Embodiments of other aspects include a non-transitory computer-readable storage medium having instructions that, when executed by a computer, cause the computer to execute a computer-implemented method for predicting the evolution of a traffic simulation.

[0042] The present invention may be implemented in a digital electronic circuit, or in computer hardware, firmware, software, or in combinations thereof. The present invention may be implemented as a computer program or a computer program product, i.e., a computer program tangibly embodied in a non-transitory information carrier, e.g., in a machine-readable storage medium or in a propagated signal, for execution by, or to control the operation of, one or more hardware modules.

[0043] The computer program may take the form of a stand-alone program, a computer program portion, or more than one computer program, and may be written in any form of programming language including a compiled or interpreted language, and it may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a data processing environment. The computer program may be deployed to be executed on one module or on multiple modules distributed at one location or across multiple locations and interconnected by a communication network.

[0044] The method steps of the present invention may be executed by one or more programmable processors executing a computer program to perform the functions of the present invention by acting on input data to produce an output. The apparatus of the present invention may be implemented as programmed hardware or as a dedicated logic circuit including, for example, an FPGA (Field Programmable Gate Array) or an ASIC (Application-Specific Integrated Circuit).

[0045] Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. In general, a processor receives instructions and data from a read only memory or a random access memory or both. An essential element of a computer is a processor that executes instructions, coupled to one or more memory devices for storing the instructions and data.

[0046] The present invention is described with respect to specific embodiments. Other embodiments are within the scope of the following claims. For example, the steps of the present invention may be performed in a different order and still achieve desirable results.

[0047] Elements of the present invention are described using terms such as "processor" and "input device". As will be apparent to those of ordinary skill in the art, such functional terms and their equivalents may refer to portions of a system that, while spatially separated, are configured to perform the defined functions. Similarly, the same physical portion of a system may provide more than one of the defined functions. For example, separately defined means may be implemented using the same memory and / or processor as needed.

[0048] Figure 1 is an overview of a general method for predicting the evolution of traffic simulation to address the above drawbacks of known simulation techniques. In S11, the computing means implementing the general method accepts an input of traffic data including data obtained from sensors within a geographic area. In S12, the computing means executes a traffic simulation using the traffic data of the geographic area. In response to a decision to model the effect of a change to the traffic simulation, in S13, the computing means selects an intervention from a plurality of potential interventions. For example, S13 first includes changing the behavior of the simulation when the predicted behavior is notified, recognized, or properly identified within a time increment of the simulation, and S13 second includes a responsive selection of an intervention for the simulation as a countermeasure to the predicted behavior. In S14, the computing means changes the parameters of the traffic simulation using the intervention to obtain the changed simulation parameters. In S15, the computing means executes the changed traffic simulation using the changed simulation parameters, and the changed traffic simulation is configured to reflect the effect of the intervention on the traffic.

[0049] According to a general method for predicting the evolution of traffic simulation and further technologies of the present application, known simulation systems can be adapted to recognize new future scenarios earlier. The system according to an embodiment can understand insights from future scenarios, specify a new simulation setup, and select an optimal intervention as a countermeasure. The technology of the present application can adapt and evolve intentions in conventional ABMS. The technology of the present application provides a more realistic simulation tailored to real-world activities.

[0050] FIG. 2 is a schematic diagram of a modular computing architecture suitable for implementing aspects of a method for predicting the evolution of traffic simulations. System 200 may communicate with an interface layer to receive live data from sensors within a geographical area of the real world. The interface layer may also be connected to a data storage layer, and data from the sensors may be stored over time. The interface layer can also receive past data stored in the data storage layer. The data storage layer may include, for example, a NoSQL database, a SQL database, and a time series database.

[0051] As shown, system 200 may be a component of DT and / or a component of its real-time DT. DT may be configured with microscopic models, mesoscopic models, macroscopic models, or any other model granularity. As an example, a microscopic model describes the behavior of vehicles and pedestrians in a traffic stream in great detail and individually. A mesoscopic model does not distinguish or track individual vehicles or pedestrians, but can represent the probability that a given vehicle or pedestrian exists at a given location, time, and speed. A macroscopic model can represent traffic at a high level of aggregation without distinguishing individual vehicles or pedestrians.

[0052] System 200 has the following components that provide the following functions in the embodiment. The Smart DT SimSights module 210 (or insights module) can play a role in providing insights from sensors and data storage layers that provide real-world data and past data. These insights provide information on predicted values and abnormal behavior patterns. For this purpose, the insights module 210 obtains information from sensors placed in the real world, from information stored in the persistence layer (data storage), and in the form of simulation information from the SimDeco module 230. The insights module 210 generates data insights. The data insights are collected and aggregated as practical insights for the purpose of providing relevant information to the SimDeco module 230 and the dynamic intervention module 220.

[0053] The dynamic intervention module 220 (or intervention module) generates potential interventions to be applied to the simulation based on any practical insights received and based on the simulator specifications. The intervention module 220 generates or determines the scope of the received data insights based on set goals such as traffic reduction, emissions reduction, etc. The scope in this context refers to the boundary or range covered by any insight within the simulation. For example, the scope may be a sub-region of interest of the geographical area being simulated, a time of interest within the simulation, and a specific traffic modality of interest. The intervention module 220 then selects the optimal intervention to be applied to the simulation based on the set goals. The intervention module 220 then generates an intervention setup assuming all the inputs received (i.e., parameters suitable for performing a simulation with interventions). The intervention module 220 then supplies any intervention setup to the SimDeco module 230.

[0054] The SimDeco module 230 (or simulation decoder module) enables the communication of a new setup to the simulation means as a practical insight or an insight to be executed, thereby changing the behavior of the original simulation. The simulation decoder module 230 changes the technical specifications of the simulation to implement the new setup and instructions. The simulator decoder module 230 transmits any simulation and / or intervention parameters or information for execution by the simulation system.

[0055] Furthermore, the simulation system may fetch data from the data storage layer for use in the simulation process. The simulation system may also store data in the data storage layer. For example, the simulation system may output simulated traffic data to be stored for future reference. Both the simulation system and the data storage layer may communicate with the interface layer to acquire and store sensor data.

[0056] System 200 can understand the predicted traffic insights, define the scope of the predicted insights, and select potential interventions. Considering the set of potential interventions, system 200 measures the impact of the interventions and selects the optimal intervention(s).

[0057] System 200 may be configured to output data or instructions to the real world. For example, when an intervention that (by simulation) indicates that a goal (e.g., congestion mitigation or emissions reduction) is satisfied is selected, System 200 may output instructions corresponding to the selected intervention for implementing the selected intervention within the actual transportation system. This may be, for example, instructions to a user to block a road or apply a contraflow arrangement. The instructions may be output in the form of computer-readable instructions so as to implement the selected intervention without human intervention (e.g., close an electronically controlled gate or display a modified speed limit on a smart highway sign).

[0058] FIG. 3 is another schematic diagram of a modular computing architecture suitable for implementing aspects of a method for predicting the evolution of traffic simulations. This schematic diagram presents the functional stages in a logical flow diagram including sub-modules for ease of understanding.

[0059] Insight module 210 receives insights from real-world sensors and any past data. Using event detection sub-module 212, insight module 210 detects potential upcoming events and, using prediction sub-module 214, predicts any new behavior. Anomaly detection sub-module 216 detects abnormal behavior (excess predicted traffic volume relative to a predetermined maximum threshold) in any of the predicted behaviors.

[0060] Decision sub-module 240 determines whether any detected abnormal traffic behavior requires mitigation by the application of an intervention.

[0061] When the decision sub-module 240 determines that an intervention needs to be applied, the control is passed to the intervention module 220. The intervention module 220 generates potential interventions based on the input received from the insight module 210. For this purpose, the intervention module 220 generates a scope of the received data insights and uses the intervention selection sub-module 224 to select the most appropriate intervention to be applied based on specific goals. The intervention definition sub-module 226 generates an intervention setup (parameters) assuming any received input, which can then be sent to the simulation decoder module 230.

[0062] When the decision sub-module 240 determines that an intervention need not be applied (e.g., when the expected effect of the detected abnormal behavior in traffic is minimal), the control can pass to the simulation decoder module 230, which may still change the underlying simulation parameters considering the predicted behavior (even in the absence of an intervention).

[0063] The simulation decoder module 230 enables the communication of new or changed simulation parameters either as practical insights (e.g., based on the predicted behavior) or as an intervention. Using the new simulation setup module 232, the behavior of the simulation can be changed considering the technical use of the simulation system in use. The realistic simulation sub-module 234 transmits the details of the new instructions and simulation parameters for execution by the simulation system.

[0064] Figure 4 is another schematic diagram of a modular computing architecture suitable for implementing aspects of a method for predicting the evolution of traffic simulations. The insight module 210, the intervention module 220, and the simulator decoder module 230 share the same functions as described above and are not fully repeated here for that reason.

[0065] The insight module 210 collects information from sensors deployed in the real world. To achieve this, the insight module 210 receives information in the form of a DT-Mob footprint that provides all the necessary information from the DT. The DT-Mob footprint is received from the DT of a geographical area and represents data that may include any of the number of vehicles, vehicle types, speed, average speed, maximum speed, minimum speed, level of congestion, number of traffic jams, level of road network usage, maximum transport capacity, weather conditions, presence of pre-determined events (e.g., concerts, sports events). Each of the data may be provided as a time series.

[0066] The sensor data is processed and a set of practical insights is supplied as output. These practical insights provide information on new or abnormal upcoming events. The practical insights provide information on the predicted behavior of mobility and information on abnormal patterns to be detected in the near future. The insight module 210 may be communicatively coupled to a persistence layer where data and operations are stored along with pre-processed information.

[0067] Any system capable of acquiring sensor data and performing traffic prediction and anomaly detection is suitable for implementation as the insight module 210. For representative examples of suitable systems, see European Patent Application No. 23382595.9 and European Patent Application No. 23382594.2.

[0068] The intervention module 220 generates potential interventions that are dynamically applied in the simulation based on practical insights and based on the DT-Mob footprint. The scope definition sub-module 222 defines the scope affected by a given practical insight.

[0069] Figure 5 is a schematic diagram of the scope definition sub-module 222. The scope may be defined using individual components for modality, time scale, and region for a given practical insight and DT-Mob footprint.

[0070] The area definition can be supplemented using a Knowledge Base (KB), and the area information can be standardized as necessary. For example, predefined sub-areas can be stored as LSOA (lower layer super output areas) or TAZ (traffic analysis zones).

[0071] The output of the individual components of the scope definition sub-module 222 may be used as input to the scope definition component, and the scope definition can be quantified based on various factors including the type of transportation, the area affected, and / or the period.

[0072] Figure 6 provides an example scope definition component and includes pseudocode showing its functionality. For the input area, period, and modality, the scope can be defined by comparing the standard area (given in the form of polygon coordinates) with the given area and by the period and traffic modality of interest.

[0073] The scope may be output as a machine-readable data file (e.g., an XML file) as shown, including individual elements for the area of the scope, the period of the scope, the modality of the scope, and the impact of the scope (such as the expected increase in the activity of the modality within the area and period). Other example impacts are represented, including a decrease peak (e.g., the activity is expected to decline before increasing), an increase peak (e.g., the activity is expected to increase before declining), a gentle increase (e.g., the activity is expected to increase by more than 0% and less than 50%), a gentle decrease (e.g., the activity is expected to decline by more than 0% and less than 50%), a significant increase (e.g., the activity is expected to increase by more than 50%), and a significant decrease (e.g., the activity is expected to decline by more than 50%). Of course, other definitions of the impact of the scope are possible.

[0074] The intervention selector sub-module 224 of the intervention module 220 selects the optimal intervention or combination of interventions, taking into account the scope and practical insights determined through the optimization of some objective function.

[0075] FIG. 7 is a schematic diagram of the intervention selector sub-module 224. The intervention logic component may be communicatively coupled to the KB, which can store pre-stored potential intervention definitions. Each potential intervention entry in the KB provides information regarding a label defining the intervention, the area to which the intervention can be applied, the period during which the intervention can be applied, the transport modality to which the intervention can be applied, and the scope (impact) of the intervention.

[0076] FIG. 8 provides an example of an intervention logic component and includes pseudocode showing its functionality. In practice, the intervention logic component filters all the stored potential interventions to identify potential interventions that are optimal given the scope. For example, considering a modality element that restricts the scope to private cars, only interventions related to private cars may be returned (e.g., excluding interventions dedicated to pedestrians).

[0077] All (filtered) potential interventions are sent to an intervention ranking component, and the interventions are scored with respect to their expected impact on the traffic simulation. This score is determined by optimizing an objective function or variable. The objective function can represent, for example, the level of traffic congestion, a monetary value associated with traffic flow (e.g., related to the fuel cost of individual drivers), the level of pollutant emissions (e.g., from all vehicles within a geographic area within the scope), or a time value associated with traffic flow (e.g., the average commuting time of all vehicles).

[0078] Figure 9 provides an example of an intervention ranking component and includes pseudocode showing its functionality. The underlying algorithm reads out all (filtered) potential interventions and solves or optimizes a multi-objective optimization problem (Pareto optimization problem) with one objective for each objective function. Equation 1 provides the general structure of such a multi-objective optimization problem: [Table 1] Here, f i (x) is the objective function for objective i, k is the number of objectives, and X ⊆ R n indicates that the problem is an n-dimensional problem. Of course, the optimization problem may instead be formulated as a maximization problem (rather than minimization).

[0079] For the application of the optimization function, predicted values, parameters required for the application of the intervention, and defined goals can be used with the aim of optimizing the selected objective function. The optimization function provides a set of scores indicating the overall expected impact of each potential intervention. The intervention ranking component can score simple independent interventions, for example, apply weights to the scores of individual interventions to define the range of combinations of multiple interventions.

[0080] Considering a multi-objective optimization problem, one can consider a set of equations in which there are sets of variables that are weighted or scored differently depending on the set goals (e.g., reduction of emissions, reduction of commuting time). For example, the scores may be different when the goals are set to "reduction of traffic volume", "maximization of the number of vehicles within a traffic-jam-free road", "reduction of pollutants", or "reduction of travel time". The elements of the objective function (x in Equation 1 above) may be individual vehicle modalities (car, van, etc.), speed for each vehicle type, number of passengers, etc. The behavior is represented as a set of functions, the aim of which is to maximize the score (alternatively, to minimize the score depending on the set goals).

[0081] A single-objective optimization problem is also possible in this context, but with a single objective function, only the optimization of a single goal can be performed. In a multi-objective optimization problem, all goals can be optimized.

[0082] The intervention selector component filters and selects the optimal intervention(s) based on the set of impact scores. For example, the configuration parameter may be configured to cause the implemented computing means to select the intervention with the highest score (or, similarly, the intervention with the lowest score if a low score indicates a positive expected effect of the intervention). Naturally, the computing means may be configured to select multiple interventions.

[0083] The intervention definition sub-module 226 of the intervention module 220 generates a setup or configuration of the intervention with sufficient detail to be interpretable by the simulation system.

[0084] FIG. 10 is a schematic diagram of the intervention definition sub-module 226. As input, the intervention definition sub-module 226 accepts the selected intervention and optionally accepts a scope definition, any practical insights, and a DT-Mob footprint. In addition to these sources, the intervention definition sub-module 226 may accept input of the specifications of the simulation system being used (e.g., an indication that the simulation is a microscopic simulation, uses 500 agents, is configured to simulate traffic, and uses an OpenStreetMap (OSM) map).

[0085] The Intervention Definition Submodule 226 constructs general information in the form of a pseudo-simulation specification. The pseudo-simulation specification is configured to supplement the specific functions of the simulation setup, and includes, for example, the type of the simulation target, the baseline of the number and speed of the simulation targets, the updated values of the number and speed of the simulation targets, the target, etc. Thus, the Intervention Definition Engine Component generates a new simulation specification or configuration based on the intervention target, the scope of the intervention, the potential future values, and the current (baseline) values.

[0086] The Intervention Definition Engine Component outputs a pseudo-simulation specification that is ready for implementation by the Simulation Decoder Module 230.

[0087] FIG. 11 is a schematic diagram of the Simulation Decoder Module 230. The Simulation Decoder Module 230 enables the communication of the pseudo-simulation specification to the simulation system. The Simulator Decoder Component of the Simulation Decoder Module 230 accepts as input the pseudo-simulation specification and optionally the DT-Mob footprint, and decodes or converts the pseudo-simulation specification into parameters (“new simulation setup”) suitable for implementation by the simulation system in use.

[0088] The Simulator Decoder Component may incorporate other additional parameters required for the setup of the simulation system, such as the network used for the simulation, agent details, etc.

[0089] The exact format and structure of the new simulator setup will vary depending on the simulation system used (in this example, a "GENERIC SIMULATOR" is being used). Examples of suitable simulation systems include proprietary software AnyLogic (developed by The AnyLogic Company), the open-source traffic modeling and simulation toolbox TRANSIMS, the open-source microscopic and continuous multi-modal simulation package SUMO (Simulation of Urban Mobility) (developed by the German Aerospace Center), FlameGPU (developed by Richmond, P. et al., doi: 10.5281 / ZENODO.5428984), proprietary software Junctions (developed by the Transport Research Laboratory), and others. Those skilled in the art will recognize that there are numerous suitable simulation systems available depending on the purpose.

[0090] As can be seen from this example, a data processing apparatus implementing a method for predicting the evolution of traffic simulation may be configured with respect to functional units or modules. Such units or modules may have a processing circuit configured to perform the roles of the above-mentioned insight module 210, intervention module 220, simulation decoder module 230, and their sub-modules and components. System 200 may be configured as an application programming interface that enables communication at least between the real-world sensors and the selected simulation system.

[0091] In one example, system 200 may be implemented using a combination of the object-oriented programming language Phython (e.g., used in multi-objective optimization processes) and Java (registered trademark) (for backend logic including interface connection with any KB).

[0092] Figure 12 is a comparison between a conventional ABMS (upper side) and an ABMS supplemented with the technology according to the embodiment (lower side). The conventional ABMS receives sensor data supplied from the real world into the simulation system and reproduces the behavior of agents within an agent-based model. As an example, the agent-based model may then generate a simulation that is 2 hours ahead of actual time with relatively low levels of accuracy and precision.

[0093] The ABMS supplemented with the technology of the present application, instead, receives sensor data and performs event detection to arrive at traffic predictions. In this example, the prediction is 2 hours ahead of actual time, and since this process can be executed faster than real time, there is no delay in providing the recommended interventions. The simulation system is then updated with information regarding the selected intervention and reproduces the changed behavior within the agent-based model. The changed agent-based model then generates a simulation that is 2 hours ahead of actual time (corresponding to the time frame used for intervention selection) with relatively high levels of accuracy and precision.

[0094] Figure 13 is a block diagram of a computing device, such as a data processing server, that can be used to implement the present invention and implement a method for predicting the evolution of traffic simulations. The computing device includes a processor 993 and a memory 994. Optionally, the computing device also includes a network interface 997 for communication with other computing devices, such as other data storage devices or external data processing devices.

[0095] Optionally, the computing device also includes one or more input mechanisms 996, such as a keyboard and a mouse, and a display unit 995, such as one or more monitors or screens. The components can be connected to each other via a bus 992.

[0096] Memory 994 may include a computer-readable medium, which term can refer to a single medium or multiple media (e.g., a centralized or distributed database and / or associated cache and server) configured to carry computer-executable instructions or store data structures. The computer-executable instructions can include, for example, instructions and data that are accessible to a general-purpose computer, a dedicated computer, or a dedicated processing device (e.g., one or more processors) and that cause such a computer to perform one or more functions or operations. Thus, the term "computer-readable storage medium" can include any medium that has the capacity to store, encode, or carry a set of instructions for machine execution and cause a machine to perform any one or more of the methods of the present disclosure. The term "computer-readable storage medium" can accordingly be understood to include, but is not limited to, solid-state memory, optical media, and magnetic media. By way of example, and not limitation, such computer-readable media can include non-transitory computer-readable storage media including random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory devices (e.g., solid-state memory devices).

[0097] Processor 993 controls the computing device and is configured to perform processing operations, such as executing computer program code stored in the memory, to implement the method for predicting the evolution of traffic simulations described herein and in the claims and related simulations. Memory 994 stores data that can be read from and written to by processor 993. As referred to herein, the processor may include one or more general-purpose processing devices such as a microprocessor, a central processing unit, etc. The processor may include a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or a processor implementing another instruction set or a combination of instruction sets. The processor may also include one or more dedicated processing devices such as an ASIC, an FPGA, a digital signal processor (DSP), a network processor, etc. In one or more embodiments, the processor is configured to execute instructions for performing the operations and steps discussed herein.

[0098] Display 995 may display a user interface. Input 996 may take the form of a touch screen or a screen and / or a keyboard and / or voice and may be used for user input. The user interface is embodied as a user application shown on the display and may optionally be connected to the audio input / output of the user device for voice input and audio output. Local storage, such as for simulation parameters or for a given potential intervention, may be provided by memory 994, and processor 993 may perform background functions. Core functions (e.g., selection of an intervention) may desirably be implemented remotely from the user device, e.g., on the cloud.

[0099] The network interface (Network I / F) 997 may be connected to a network such as the Internet and can be connected to other such computing via the network. The Network I / F 997 can control data input from other devices via the network or data output to other devices. Other peripheral devices such as microphones, speakers, etc. may be included in the computing device.

[0100] The insight module 210, a processor 993 (or multiple processors 993) that executes instructions stored in a part of the memory 994 and exchanges data via the network I / F 997. In particular, the processor 993 executes processing instructions to extract traffic data from sensors and identify and analyze any upcoming events in the traffic system that may require intervention.

[0101] The intervention module 220, a processor 993 (or multiple processors 993) that executes instructions stored in a part of the memory 994 and exchanges data via the network I / F 997. In particular, the processor 993 executes processing instructions to determine the scope of the traffic simulation to be intervened in and select the intervention(s) (including multiple).

[0102] The simulation decoder module 230, a processor 993 (or multiple processors 993) that executes instructions stored in a part of the memory 994 and exchanges data via the network I / F 997. In particular, the processor 993 executes processing instructions to interpret the intervention parameters within the data structure readable by a given simulation system.

[0103] The processor 993 that embodies the insight module 210, the intervention module 220, and the simulation decoder module 230 can execute instructions stored in a part of the memory 994 to exchange data between the modules via the bus 992 and with external entities such as traffic sensors via the network I / F 997.

[0104] The method for implementing the present invention may be executed on a computing device such as that shown in FIG. 13. Such a computing device need not have all the components shown in FIG. 13 and may consist of a part of these components. The method for implementing the present invention may be executed by a single computing device that communicates with one or more data storage servers via a network. The computing device may itself be a data storage that stores pre-set interventions, buffered sensor data, past sensor data, a multi-function optimization model, and parameters, etc.

[0105] The method for implementing the present invention may be executed by a plurality of computing devices that cooperate with each other. One or more of the plurality of computing devices may be a data storage server that stores at least a part of pre-set interventions, buffered sensor data, past sensor data, a multi-function optimization model, and parameters, etc.

[0106] The various methods described above may be implemented by a computer program. The computer program may include computer code (e.g., instructions) arranged to instruct a computer to perform one or more functions of the various methods described above. For example, the steps of the method described with respect to FIG. 1 may be executed by computer code. The steps of the methods described above may be executed in any suitable order. A computer program and / or code for performing such a method may be provided to a device such as a computer on one or more computer-readable media, or more generally, in a computer program product. The computer-readable media may be transient or non-transient. The one or more computer-readable media can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, or a propagation medium for data transmission, for example, for downloading code via the Internet. Alternatively, the one or more computer-readable media can take the form of one or more physical computer-readable media such as semiconductor or solid state memory, magnetic tape, removable computer diskette, random access memory (RAM), read-only memory (ROM), hard magnetic disk, and optical disks such as CD-ROM, CD-R / W or DVD. The instructions can also be fully or at least partially present in the memory 994 and / or in the controller circuitry of the processor 993 during its execution by a computing system, and the memory 994 and the controller circuitry of the processor 993 also constitute a computer-readable storage medium.

[0107] In practice, the modules, components, and other features described herein may be implemented as discrete components or incorporated into the functionality of hardware components such as ASICs, FPGAs, DSPs, or similar devices.

[0108] A "hardware component" is a tangible (e.g., non-transitory) physical component (e.g., a set of one or more processors) capable of performing a particular operation and may be configured or arranged in a particular physical manner. A hardware component may include dedicated circuitry or logic that is permanently configured to perform a particular operation. A hardware component may have a dedicated processor such as an FPGA or ASIC. A hardware component may also include programmable logic or circuitry that is temporarily configured by software to perform a particular operation.

[0109] Furthermore, modules and components may be implemented as firmware or functional circuitry within a hardware device. Also, modules and components may be implemented in any combination of hardware devices and software components, or solely in software (e.g., code stored in a machine-readable medium or a transmission medium or otherwise embodied).

[0110] Unless otherwise specified, as will be apparent from the following discussion, throughout this specification, discussions using terms such as "receive," "determine," "compare," "enable," "hold," "identify," "acquire," "access," etc., refer to the actions and processes of a computer system or similar electronic computing device that manipulates and transforms data represented as physical (electronic) quantities within the registers and memories of the computer system into other data similarly represented as physical quantities within the computer system memory or registers, or other such information storage, transmission, or display devices.

[0111] Particular embodiments have been described, but these embodiments are presented by way of example only and are not intended to limit the scope of the invention. In fact, the novel methods and apparatuses described herein may be embodied in a variety of other forms, and furthermore, various omissions, substitutions, and changes in the form of the methods and apparatuses described herein may be made.

[0112] The following appendices are useful for understanding the present invention.

[0113] (Appendix 1) A method implemented by a computer for predicting the evolution of traffic simulation, comprising: receiving an input of traffic data including data obtained from sensors within a geographical area; executing a traffic simulation using the traffic data of the geographical area; selecting an intervention from a plurality of potential interventions in response to a determination to model the effect of a change to the traffic simulation; changing parameters of the traffic simulation using the intervention and obtaining changed simulation parameters; executing a changed traffic simulation configured to reflect the effect of the intervention on traffic using the changed simulation parameters and having.

[0114] (Appendix 2) The determination to model the effect of a change to the traffic simulation is identifying upcoming events within the traffic simulation; predicting the impact of the upcoming events on the traffic simulation; and when the impact exceeds a predetermined threshold, determining to model the effect of the change on the traffic simulation and having, the method according to Appendix 1.

[0115] (Appendix 3) Selecting the intervention from the plurality of potential interventions is determining the scope of the traffic simulation to receive the intervention; Calculating a plurality of scores corresponding to the plurality of potential interventions, the scores indicating to what extent the corresponding potential intervention affects the target variable of the traffic simulation, Selecting the potential intervention with the top score as the intervention and having The method according to Appendix 1 or 2.

[0116] (Appendix 4) Calculating the plurality of scores corresponding to the plurality of potential interventions includes executing an optimization process configured to solve an optimization problem with respect to the potential interventions. The method according to Appendix 3.

[0117] (Appendix 5) The scope of the traffic simulation to receive the intervention has any or all of traffic modalities, time scales, and sub-regions of the geographical area. The method according to Appendix 3 or 4.

[0118] (Appendix 6) The intervention is the following The number of vehicles in the geographical area Transport modalities in the geographical area The positions of vehicles in the geographical area The directions of movement of vehicles in the geographical area The speeds of vehicles in the geographical area The number of accessible lanes and / or roads The number of accessible stops and / or stations, and The operation of traffic lights is a constraint on any or all of The method according to any one of Appendices 1 to 5.

[0119] (Appendix 7) Selecting a plurality of interventions from the plurality of potential interventions The method according to any one of Appendices 1 to 6.

[0120] (Appendix 8) The effect of the intervention on traffic is as follows: The number of vehicles in the geographical area, The number of each of the plurality of transport modalities in the geographical area, The position of the vehicles in the geographical area, The direction of movement of the vehicles in the geographical area, The speed of the vehicles in the geographical area, The average speed of the vehicles in the geographical area, The minimum speed and / or maximum speed of the vehicles in the geographical area, The level of congestion in the geographical area, The number of traffic jams in the geographical area, The level of use of the road network in the geographical area, The maximum transport capacity in the geographical area, and The level of emission power in the geographical area which is any or all of the changes among: The method according to any one of Appendices 1 to 7.

[0121] (Appendix 9) Receiving the input of past traffic data of the geographical area, and further comprising executing the traffic simulation using the traffic data and the past traffic data, The method according to any one of Appendices 1 to 8.

[0122] (Appendix 10) The traffic data includes data obtained from the digital twin of the geographical area, The method according to any one of Appendices 1 to 9.

[0123] (Appendix 11) The sensor is as follows: A GPS sensor or antenna, A cellular network sensor or antenna, A loop sensor, A camera sensor, and Vehicle parking sensor Any one or all of the following, The method described in any one of Supplementary Notes 1 to 10.

[0124] (Supplementary Note 12) Changing the parameters of the traffic simulation and executing the changed traffic simulation are executed in parallel with the traffic simulation. The method described in any one of Supplementary Notes 1 to 11.

[0125] (Supplementary Note 13) A memory storing computer-executable instructions for executing the method described in any one of Supplementary Notes 1 to 12, A processor configured to execute the instructions And a data processing device having the same.

[0126] (Supplementary Note 14) A computer program having instructions, When the program is executed by a computer, the computer is caused to execute the method described in any one of Supplementary Notes 1 to 12. Computer program.

[0127] (Supplementary Note 15) A computer-readable medium having instructions that, when executed by a computer, cause the computer to execute the method described in any one of Supplementary Notes 1 to 12.

Explanation of Signs

[0128] 200 System 210 Smart DT SimSights Module (Insight Module) 212 Event Detection Sub-module 214 Prediction Sub-module 216 Abnormal Behavior Detection Sub-module 220 Dynamic Intervention Module 224 Intervention Selection Sub-module 226 Intervention Definition Sub-module 230 SimDeco Module 232 New Simulation Setup Module 234 Realistic Simulation Sub-module 240 Decision Sub-module

Claims

1. 1. A computer-implemented method for predicting the evolution of a traffic simulation, comprising the steps of: accepting an input of traffic data, the data including data obtained from sensors within a geographic region; performing a traffic simulation using the traffic data for the geographical area; selecting an intervention from a plurality of potential interventions in response to a decision to model an effect of a change to the traffic simulation; - modifying parameters of the traffic simulation using said intervention to obtain modified simulation parameters; running a modified traffic simulation with the modified simulation parameters, the modified traffic simulation being configured to reflect an effect of the intervention on traffic; The method according to claim 1,

2. The determination of modeling an effect of a change to the traffic simulation includes: identifying an upcoming event within the traffic simulation; predicting an impact of the upcoming event on the traffic simulation; and determining to model an effect of said change on said traffic simulation if said impact exceeds a predefined threshold; having The method of claim 1.

3. Selecting the intervention from the plurality of potential interventions comprises: determining an area of ​​the traffic simulation that is subject to the intervention; and calculating a number of scores corresponding to the number of potential interventions, the scores indicating how a corresponding potential intervention would affect a target variable of the traffic simulation; selecting the top scoring potential intervention as the intervention; having The method of claim 1.

4. and computing the plurality of scores corresponding to the plurality of potential interventions comprises executing an optimization process configured to solve an optimization problem with respect to the potential interventions. The method according to claim 3.

5. the scope of the traffic simulation subject to the intervention comprises any or all of a traffic modality, a time scale, and a sub-area of ​​the geographical area; The method according to claim 3.

6. The intervention comprises: The number of vehicles within the geographic area; transportation modalities within said geographic region; the location of the vehicle within said geographic region; the direction of movement of the vehicle within said geographic area; the speed of the vehicle within said geographic area; the number of accessible lanes and / or roads; The number of accessible stops and / or stations; and Traffic light operation A restriction on any or all of the following: The method of claim 1.

7. selecting a plurality of interventions from the plurality of potential interventions. The method of claim 1.

8. The effect of the intervention on traffic is: The number of vehicles within the geographic area; a respective number of a plurality of transportation modalities within said geographic region; the location of the vehicle within said geographic region; the direction of movement of the vehicle within said geographic area; the speed of the vehicle within said geographic area; the average speed of vehicles within said geographic area; minimum and / or maximum speeds of vehicles within said geographical area; the level of congestion within said geographic area; the number of traffic jams within said geographic area; the level of use of the road network within said geographic area; Maximum transportation capacity within said geographic area; and The level of emission power within said geographic area Any or all of the following changes: The method of claim 1.

9. accepting an input of historical traffic data for said geographic area; and performing the traffic simulation using the traffic data and the historical traffic data. The method of claim 1.

10. The traffic data comprises data obtained from a digital twin of the geographic area. The method of claim 1.

11. The sensor includes: GPS sensor or antenna, A cellular network sensor or antenna; Loop Sensor, Camera sensor, and Vehicle Parking Sensor Any or all of the following: The method of claim 1.

12. modifying parameters of the traffic simulation and running the modified traffic simulation are performed in parallel with the traffic simulation. The method of claim 1.

13. a memory storing computer executable instructions; Executing the computer-executable instructions Accepting an input of traffic data, the traffic data including data obtained from sensors within a geographic region; performing a traffic simulation using said traffic data for said geographical region; selecting an intervention from a plurality of potential interventions in response to a decision to model an effect of a change to the traffic simulation; modifying parameters of the traffic simulation using said intervention to obtain modified simulation parameters; running a modified traffic simulation with the modified simulation parameters configured to reflect an effect of the intervention on traffic. and a processor configured to A data processing device having:

14. 1. A computer program having instructions, comprising: The instructions, when the computer program is executed by a computer, cause the computer to: accepting an input of traffic data including data obtained from sensors within a geographic region; performing a traffic simulation using said traffic data for said geographical region; selecting an intervention from a plurality of potential interventions in response to a decision to model an effect of a change to the traffic simulation; modifying parameters of the traffic simulation using said intervention and obtaining modified simulation parameters; running a modified traffic simulation with the modified simulation parameters, the modified traffic simulation being configured to reflect an effect of the intervention on traffic. Computer program.

15. When executed by a computer, the computer accepting an input of traffic data including data obtained from sensors within a geographic region; performing a traffic simulation using said traffic data for said geographical region; selecting an intervention from a plurality of potential interventions in response to a decision to model an effect of a change to the traffic simulation; modifying parameters of the traffic simulation using said intervention and obtaining modified simulation parameters; running a modified traffic simulation with the modified simulation parameters, the modified traffic simulation being configured to reflect an effect of the intervention on traffic. A computer readable medium having instructions.

Images