EP4416660A1

EP4416660A1 - Control of a complex environment

Info

Publication number: EP4416660A1
Application number: EP22801098.9A
Authority: EP
Inventors: Jonas Christoffer VILLUMSEN
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2021-10-14
Filing date: 2022-10-10
Publication date: 2024-08-21
Also published as: WO2023061959A1

Abstract

A computer-implemented method is provided for controlling a complex environment, wherein the environment has plural access points for entrance and exit of people into and out of the environment, fixed infrastructure elements through which people move on journeys through the environment between the access points, and variable resources by which movement of people through and between the fixed infrastructure elements and the access points can be controlled. The method includes steps of: predicting the numbers of people who will enter the environment and exit the environment at each of the access points over a time period at a predetermined time in the future; optimising an objective function to determine a configuration of the variable resources which will enable the predicted numbers of people to enter the environment and exit the environment at the respective access points at the predetermined time; and controlling the variable resources to implement the determined configuration by the predetermined time.

Description

CONTROL OF A COMPLEX ENVIRONMENT

This application claims priority from EP 21202701.5 filed 14 October 2021 , the contents and elements of which are herein incorporated by reference for all purposes.

Field of the Invention

The present invention relates to control of complex environments.

Background

In many large, complex environments, such as train stations, ports, hospitals, airports, shopping malls, and office buildings, facilitating the movement of pedestrians constitutes one of the critical functions of the space encompassed by the environment. To this end, various types of equipment may be installed to manage and direct the flow of people, including escalators, elevators, gates, information displays and signs, lighting, automatic doors, self-driving vehicles, etc.

The congregation of large volumes of people in a limited physical space may lead to congestion, disease spread, discomfort, and excessive time spent in transit. Aging and inadequately dimensioned infrastructure leads to significant issues in dealing with increasing volumes of people flow, while the cost of expanding existing infrastructure to accommodate increasing volumes are often very high, if expansion is even feasible at all. Simply alleviating bottlenecks in one place, may lead to congestion arising in other places. Hence, it is desirable to take a holistic view of the entire connected space of the environment.

Pedestrian flow simulation software, such as AnyLogic, OpenSpace and MassMotion, can provide a detailed simulation of individual pedestrian behaviour and their movement e.g. through a building or a space, and can give valuable insights into the flow of people given a specific deployment and configuration of equipment and infrastructure. However, such software is generally not suited to finding a best configuration of possible infrastructure resources for tactical and operational control subject to operational constraints.

US 2020/0114935 proposes a method to dynamically deploy platform bridges to allow smooth means of transfer of passengers among multiple platforms in a station. US 2018/0039949 proposes a method to optimise and synchronise people flows by scheduling individual users' activities and activity sequences to resources resulting in recommended start and end times for each activity and/or user. US 2019/0212719 proposes a method to control equipment to attain an ideal situation based on sensor input such as congestion level.

The present invention has been devised in light of the above considerations. Summary of the Invention

The current invention seeks to modulate the flow of people within a complex environment to attain improved operational outcomes. This can be achieved by changing the configuration and allocation of equipment, spaces, staff, and other resources, such that a pre-defined objective function is optimised. The objective function may include one or more parameters relating to different operational considerations, e.g. congestion levels, total travel time, contraflows, operational costs, energy consumption, etc., depending on requirements of the environment. In particular, the objective function may include one or more parameters relating to respective physical operational considerations, such as any one or more of congestion levels, total travel time, contraflows, and energy consumption.

Thus, in general terms the present invention provides a computer-implemented method for controlling a complex environment (which is typically a complex built environment), wherein the environment has plural access points for entrance and exit of people into and out of the environment, fixed infrastructure elements through which people move on journeys through the environment between the access points, and variable resources by which movement of people through and between the fixed infrastructure elements and the access points can be controlled, the method including steps of: predicting the numbers of people who will enter the environment and exit the environment at each of the access points over a time period at a predetermined time in the future; optimising an objective function to determine a configuration of the variable resources which will enable the predicted numbers of people to enter the environment and exit the environment at the respective access points at the predetermined time; and controlling the variable resources to implement the determined configuration by the predetermined time.

The variable resources can be any one or more physical variable resources selected from: escalators, lifts, stairways, moving walkways, corridors, lanes, barriers and gates. Additionally or alternatively, the variable resources can be any one or more user interface variable resources selected from: signage, displays and announcements. The variable resources can be fixed or mobile. Staff can also be categorised as user interface variable resources, e.g. to the extent that the availability of staff and/or staff actions can impact on movement of people.

Typically, the greater the complexity of the environment, the higher the number of access points the environment will have. Thus the complex environment may have at least three, four, five, six, eight, twelve or twenty of the access points. The access points may be separated from other access points of the complex environment by fixed infrastructure elements and/or variable resources. However, the complex environment may have a subset of least three of the access points which are each separated from the other access points of the subset by variable resources (i.e. any journey between any pair of these access points requires people to traverse one or more variable resources), whereby movement of people between the access points of the subset can be controlled. Such access points may be termed “variable resource-separated” access points. Typically, the greater the complexity of the environment, the higher the number of such “variable resource-separated” access points the environment will have. For example, the complex environment may have at least three, four, five, six, seven, eight, twelve or twenty “variable resource-separated” access points.

More specifically, in a first aspect, the present invention provides a computer-implemented method for controlling a complex environment, wherein the environment has plural access points for entrance and exit of people into and out of the environment, fixed infrastructure elements through which people move on journeys through the environment between the access points, and variable resources by which movement of people through and between the fixed infrastructure elements and the access points can be controlled, the method including steps of: detecting the numbers of people entering the environment and exiting the environment at each of the access points over a time period; predicting, on the basis of the detected numbers of people, the numbers of people who will enter the environment and exit the environment at each of the access points in a similar time period at a predetermined time in the future; optimising an objective function to determine a configuration of the variable resources which will enable the predicted numbers of people to enter the environment and exit the environment at each of the access points at the predetermined time; and controlling the variable resources to implement the determined configuration by the predetermined time.

Thus, advantageously, the method can predict future people flows in the so that the variable resources, and particularly physical variable resources, can be reconfigured or reallocated in advance to meet expected demands, rather than merely reacting to changes in demand when they occur.

The detecting of the numbers of people can be performed, for example, by video, Lidar, infra-red sensors, NFC, bluetooth sensors etc.

The prediction of the numbers of people who will enter the environment and exit the environment at each of the access points over the similar time period at the predetermined time in the future may also be based on stored historical data of numbers of people entering the environment and exiting the environment at each of the access points in a similar time period at an equivalent time or times in the past. In this way the prediction accuracy can be improved.

The prediction of the numbers of people who will enter the environment and exit the environment at each of the access points over the similar time period at the predetermined time in the future may also be based on external variables. Again, in this way the prediction accuracy can be improved. As examples, the external variables can include current local weather conditions, current local transport conditions (e.g. derived from train management systems, TMSs) and/or holiday information. The prediction step can also include predicting, for each access point acting as an origin of people entering the environment over the similar time period at the predetermined time the destination access points at which those people will exit the environment, and also predicting for each destination access point of an origin access point the respective fraction of the number of people entering that origin access point who will exit at that destination access point. These predictions can be generated in the form of an origin/destination matrix which provides for the predetermined time in the future and for each origin/destination pair the respective fraction of the number of people entering that origin access point who will exit at that destination access point. In the origin/destination matrix, each fraction can be specified in more detail by allocating respective parts of the fraction to respective people classifications (discussed below). In the optimisation step, the optimisation of the objective function can then determine a configuration of the variable resources which will enable the people entering each origin access point to travel to its respective predicted destination access points in numbers corresponding to the respective predicted fractions. This refinement of the prediction step and its use in the optimisation step can improve the ability of the method to meet future demands. The predictions of the destination access points and the respective fractions can be based on stored historical data of destination access points and respective fractions at an equivalent time or times in the past. In this way the prediction accuracy can be improved. Conveniently, the detection step can include detecting the destination access points of people entering the environment at each origin access point over the time period; and the predictions of the destination access points and the respective fractions can then be based on the detected destination access points and respective fractions. For example, the detecting of the destination access points can be performed by obtaining electronic travel card data, ticketing data, wifi tracking data etc.

The detection step can include detecting people classifications, such as pedestrians, wheelchair users, and persons with buggies, and detecting the numbers of people entering the environment and exiting the environment at each of the access points over the time period in each classification. The prediction step can then be performed for each detected people classification, and in the optimisation step, the optimisation of the objective function can determine a configuration of the variable resource which is enabling for all the detected classifications. In this way, different types of user may not be disadvantaged by a configuration of the variable resources which does not meet their requirements.

The detecting, predicting, optimising and controlling steps are typically repeated at intervals to update the configuration of the variable resources. Thus the resources can be continuously reallocated and reconfigured as needed.

The complex environment may be a train station, a port, a hospital, an airport, a shopping mall, an office building, a theme park, a ferry or cruise ship, a campus, or a conference venue.

The method is computer-implemented. Accordingly, further aspects of the present disclosure provide: a computer program comprising code which, when the code is executed on a computer, causes the computer to perform the method of the first aspect; a computer readable medium storing a computer program comprising code which, when the code is executed on a computer, causes the computer to perform the method of the first aspect; and a computer system programmed to perform the method of the first aspect. In yet another aspect, the present disclosure provides a complex environment controlled by the computer system of the previous aspect, the complex environment having plural access points for entrance and exit of people into and out of the environment, fixed infrastructure elements through which people move on journeys through the environment between the access points, and variable resources by which movement of people through and between the fixed infrastructure elements and the access points can be controlled. The complex environment may have sensors for detecting the numbers of people entering the environment and exiting the environment at each of the access points over a time period.

The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

Summary of the Figures

Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:

Figure 1 shows a schematic architecture of a computer system for performing a method of controlling a complex environment,

Figure 2 shows a representation of the complex environment as a network;

Figure 3 is a flow chart describing data flow and process steps in the method for controlling the complex environment; and

Figure 4 shows schematically information sharing in the method.

Detailed Description of the Invention

Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

Figure 1 shows a schematic architecture of a computer system for performing a method of controlling a complex environment, such as a train station, a port, a hospital, an airport, a shopping mall, an office building, a theme park, a ferry or cruise ship, a campus, or a conference venue.

The complex environment has plural access points for entrance and exit of people into and out of the environment, fixed infrastructure elements through which people move on journeys through the environment between the access points, and variable resources by which movement of people through and between the fixed infrastructure elements and the access points can be controlled. In the example of a train station, the access points can be pedestrian entrances to the station from the outside, and platforms at which passengers embark onto and disembark from trains, the fixed infrastructure elements can be concourses, passageways, hallways, tunnels, stairways, waiting areas etc., and the variable resources such as escalators, lifts, stairways, moving walkways, corridors, lanes, barriers, gates, signage, displays, announcements etc. They can be fixed or mobile. Staff can also be categorised as variable resources to the extent that the availability of staff and/or staff actions can impact on movement of people. Figure 2 shows a representation of the complex environment as a network in which edges (links) represent connecting infrastructure such as concourses, passageways, escalators, shuttles etc., some of which will be variable resources or fixed infrastructure elements incorporating variable resources, and others of which will be fixed infrastructure elements without any variable resources, and nodes represent logical connections or access points. Purely, as an example, access points are indicated on Figure 2 as nodes A’ or A, and connecting infrastructure elements are indicated as edges. More specifically, fixed infrastructure elements without any variable resources are indicated on Figure 2 as edges F, and variable resources or fixed infrastructure elements incorporating variable resources are indicated on Figure 2 as edges V. The environment has six access points, of which five form a subset of “variable resource- separated” access points A’ because they are separated from the other access points A' of the subset by variable resources V (i.e. any journey between any pair of these access points requires people to traverse one or more edges V). The sixth access point A is not part of this subset because it is connected to the adjacent access point A’ by a path formed solely of edges F.

In the Figure 2 representation, the abovementioned sixth access point A and adjacent access point A’ are surrounded by a dashed outline. The respective designations of these two access points could be reversed (i.e. so that the one that is indicated as A’ is indicated as A, and the one that is indicated as A is indicated as A’), but they could not both be type A or both be type A’. To an extent these two access points can be considered as acting as a combined, single access point because the edges F which connect them do not permit control of the movement of people therebetween. However, for other reasons, it is generally beneficial to treat these two access points separately in the representation of Figure 2. In particular, by treating them separately it is possible for the system to account for and to predict different people inflows and outflows at the access points, which may in turn impact on how people flows are predicted and controlled elsewhere in the complex environment. The computer system can manage pedestrian flows in such complex environments with high throughputs of pedestrian volumes. As demand for capacity at various locations across the environment changes throughout the day, week, and year, the deployment and configuration of equipment, spaces, and staff can be controlled to accommodate these changes in an automated and/or semi-automated fashion. Significantly, the system predicts people flows at a predetermined time in the future so that variable resources can be reconfigured or reallocated in advance to meet expected demands, rather than merely reacting to changes in demand when they occur.

As an example, a train or metro station may have access points which are pedestrian entrances/exits, and platforms for embarkation/disembarkation of passengers. The station may have an opportunity to automatically control:

• escalator directions, the opening, closing, and direction of automatic barriers, the direction of flow in narrow corridors by making them uni-directional, associated signage and announcements.

As another example, airports may have an opportunity to automatically control:

• escalator and travelator directions,

• the opening, closing, and direction of automatic gates at border control,

• the deployment of automated shuttles e.g. driverless trains and busses,

• the direction of flow in narrow corridors by making them uni-directional,

• the direction of flow through automated entry/exit doors,

• the assignment of automated self-check-in points to airlines,

• associated signage and announcements.

Automated controls may be supplemented by decision support for semi-automated decisions for further resource allocation, where manual assistance is required, e.g.:

• opening and closing of lanes in security,

• opening and closing of check-in counters,

• re-assignment of aircrafts to gates.

The system can seek to reduce queue lengths, transit walking times, and resource energy consumption by finding a best possible combination of resource assignments and configurations, considering the flow of passengers across the entire airport from entry to gate and vice versa. The optimal configuration exemplified by an allocation plan of variable resources can automatically adapt to changes in demand based on inputs from sensors and external variables.

Returning to Figure 1, the computer system has a central unit containing a sensor data pre-processor 201 , a prediction module 202, an origin and destination matrix generator 203 and an optimisation module 204. The central unit communicates with a database 205 and has a control interface 208 for user interaction with the unit. Via a network 100, the central unit receives information from sensors 110 and an external variable module 111. It also sends control commands to equipment 115, information signs 116 and other devices 117 to control the variable resources of the environment. The operation of these features of the system will now be described in relation to Figure 3, which is a flow chart describing data flow and process steps in relation to elements of the system in the method for controlling the complex environment.

The sensors 110, which may include video, Lidar, infra-red sensors, NFC, bluetooth sensors, etc., collect information about the flow of people by monitoring the number of passengers within spatially limited zones. In particular, the sensors monitor the numbers of people entering the environment and exiting the environment at each of the access points over a time period. The volume and direction (i.e. entering or leaving) of people movement can be grouped by profiles representing different people classifications, e.g. wheel chair users, persons with buggies, commuters, leisure travellers, visually impaired travellers etc. Profiles are typically based on observable visual characteristics of people, and may be stored in the database 205.

As a result of representing the infrastructure as a logical connection of nodes and edges, as shown in Figure 2, the network can be converted into a directed graph depicting possible paths for journeys through the complex environment. In the directed graph, each node can be defined by a unique identifier and associated with optional properties, such as position and node type. In addition, each edge can be defined by a unique identifier and assigned information such as start node, end node, edge-type and resources (if the edge is or incorporates a variable resource, rather than being fixed infrastructure without a variable resource). Each edge can also contain a mapping between people classifications and their corresponding weightings for that edge. The weightings indicate the perceived cost for a particular class of people to traverse that edge. For example, the weightings may be in proportion to distance, time, or perceived difficulty of traversing the edge. Consequently, the weightings can be used as variables for evaluating the reasonability of journeys configured from edges. Alternatively, the mapping between people classifications and their corresponding weightings for an edge can be calculated dynamically based on the people classification profiles, capacities of the connecting infrastructure, and availabilities of the variable resources. For instance, if the profiles or variable resources availabilities also store a weighting for each classification, then the system can estimate the weightings applicable to each edge using this information and the predicted number of people travelling through that edge.

The edge capacity defines the maximum people flow that a given connecting infrastructure edge V or F can handle. The variable resources allocated to an edge V are chosen from a subset of variable resources allowed for that edge. Each resource allocated to an edge V may add to or subtract from the capacity of that edge as well as to the operational characteristics of the edge.

The operational characteristics of an edge V or edge F can include, for example, its capability to handle bidirectional flow. If an edge is defined as bi-directional, it allows flows in both directions (and conversely if an edge is defined as uni-directional, flow is only allowed in one direction between the nodes at either end of the edge). The operational characteristics of an edge V can also include, for example, its definition as symmetric or asymmetric. If an edge V is defined as symmetric then the variable resources allocated to the edge handle flows in both directions. Otherwise, if an edge V is defined as asymmetric, resources must be explicitly assigned to each direction separately.

Each variable resource has an associated availability indicating the number of resources available (e.g. the number of staff available for deployment in a given period, or the number of escalators in a bank of escalators). In addition, the variable resource may have a default capacity attribute which represents the flow capacity of that resource, and/or a default cost attribute which represents the operational cost of allocating that resource to an edge V.

Conveniently, an edge library may be provided which contains sets of edge parameters which define commonly used edge types. These edge parameter sets can be applied as default parameter sets, and may include whether the edge is bi-directional or uni-directional, whether it is symmetric or asymmetric, and typical weightings. If a user calls a particular edge type out of the edge library for use in building an edge of the directed graph representing a complex environment, the corresponding default parameter set is automatically assigned to the edge. Considering the system in more detail, and referring to Figure 3, the raw sensor data 601 from the sensors 110 is conveniently pre-processed by the sensor data pre-processor 201 into aggregated numeric information consisting of people count data 602 by classification, time period, location, and direction of flow, as exemplified in Table 1. In this example each time period is of ten minutes duration, and the number for each entry is the total number of people entering or exiting in that classification and at that entrance in the time period starting at the given time. The numeric passenger counts are stored in the database 205.

Table 1 : Example of people count table 602

An external variables module 111 collects external variables 603, such as local weather data, local holiday information, etc. as exemplified in Table 2, and stores the data in the database 205. These external variables can be used to help predict people flows, as explained in more detail below. In the example of a train station, current local transport conditions (e.g. obtained from train management systems, TMSs) that provide up to date information on platform departures and arrivals can also be included in the external variables 603.

Table 2: Example of external variables table 603 The external variables can be updated at the same or over a longer time periodicity as the people count data 602, e.g. depending on the typical rate and amount of change that is observed in these variables. In the example of table 2, weather data is updated at hourly intervals.

The prediction module 202 receives the people count data 602 and uses this to predict future people counts 604 over similar time periods (ten minutes in the example of Table 1) at a predetermined time in the future. To improve this prediction, the prediction module 202 can make use of historic people count data 206 of the numbers of people entering the environment and exiting the environment at each of the access points over a similar time period at an equivalent time or times in the past. Moreover, the prediction module 202 can make use of the external variables 603 to further improve the prediction. Historic external variables 207 stored in the database 205 can also be used to improve the prediction, e.g. in combination with the historic people count data 206.

The prediction module 202 can use a neural network or other machine learning approaches to draw inferences from the people count data 602 and other data (i.e. any one or more of the historic people count data 206, external variables 603, and historic external variables 207). However, another possible approach is simply to extrapolate predictions based on historical values e.g. on a weekly basis. The output of the prediction module 202 is predicted people count data 604 that can take the same format as the people count data 602 of Table 1.

Next, an O/D (origin/destination) matrix generator 203 can be used to calculate estimated passenger flow volumes by origin and destination at the predetermined time in the future. The calculation uses the predicted people counts 604 and O/D splits 605. The O/D splits can be estimated from external sources such as electronic travel card data, ticketing data, wifi tracking data, etc. The O/D matrix generator 203 outputs for each people classification, future time period, and origin/destination pair the estimated volume of people as shown in Table 3.

Table 3: Example of part of O/D matrix 606

The optimisation module 204 then receives the O/D matrix 606 along with infrastructure data 608 which specifies the existing infrastructure elements and how they are connected. These may be specified as an undirected graph, as shown in Figure 2, onto which movement of pedestrian flows can be projected. Each combination of a people classification type and an origin and destination node pair within the undirected graph can conveniently be termed a “commodity”. It may be identified by a respective ID, and in the O/D matrix has an associated demand which is the estimated volume of people. Additionally, the infrastructure data 608 can include a mapping 611 between infrastructure location and possible variable resources that may be assigned to it, as shown in Table 4. Table 4: Example of the mapping 611 of variable resource to connecting infrastructure

In Table 4 the connecting infrastructure elements and variable resources are identified by descriptive names, but more conveniently they may be identified by edge IDs and resource IDs.

Also available to the optimisation module 204, e.g. from storage in the database 205, is a characteristics table 607 defining the characteristics of different types of variable resources that may be available, and also the characteristics of certain connecting infrastructure elements, as shown in Table 5.

Table 5: Example of characteristics table 607

In addition the optimisation module 204 receives a resource availability table 610 defining the availability of resources/elements by time period, as shown in Table 6. Table 6: Example of resource availability table 610

Figure 4 shows schematically how information is shared between the O/D matrix 606, the characteristics table 607, the resource availability table 610 and the mapping 611.

The optimisation module 204 calculates the configuration and allocation of variable resources that facilitates the flow of pedestrians in the best possible way according to a predefined objective function and constraints. In one example, the model may be formulated as mixed integer program (MIP) and solved by e.g. a standard MIP solver or by a heuristic. The model may incorporate constraints on availability of resources according to the resource availability table 610 and applicability of variable resources to fixed infrastructure elements according to the mapping 611. Additional operational constraints may be defined by the user. The output of the optimisation module 204 is a resource allocation plan 609, as shown in Table 7. The resource allocation location plan can be communicated to static and mobile equipment 115 and information signs 116 for automatic deployment and configuration, as well as to staff via stationary and handheld devices 117.

Table 7: Example of resource allocation plan 609

The following is an example of a general MIP for determining an optimal resource allocation plan. min/(x) + 5(y) x,y s. t. Ax < By Cx < d Dy < e where x,y, d,e are column-vectors and A,B, C,D are matrices of appropriate dimensions, x is a decision vector of real variables and y is a decision vector of binary variables. For example, x may represent the flow of people at various edges across the network depicted in Fig. 2, while y may represent the configuration of the variable resources and auxiliary binary variables, e.g. variables to indicate points of congestion. The function /(x) can be the total distance travelled by all people, while gty) may represent the resource allocation costs (e.g. energy consumption costs) as well as a penalty for congestion. The objective function /(x) + gty) may be constructed appropriately to provide a desired balance between multiple objectives. The first set of constraints expresses direct relations between the people flow x and resource and congestion variables y, e.g. resource allocation impact on flow capacity and the impact of flow on congestion levels. The second set of constraints expresses, e.g. people flow conservation (ensuring paths from points of entry to points of exit), while the third set of constraints expresses, e.g. resource availability constraints. Other forms of the optimisation model may be used and may include different MIP formulations from the one presented here.

The control interface 208 provides a user interface to allow an operator to interact with the central unit and modify parameters and behaviours. This can include modifying the objective function, overriding the resource allocation plan 609, fixing resource allocations (e.g. fixing certain escalator directions), modifying the input and recalculating, etc.

The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and/or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and/or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example +/- 10%.

Embodiments may be described as a process which is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations may be re-arranged. A process is terminated when its operations are completed, but could have additional steps not included in the figure. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination corresponds to a return of the function to the calling function or the main function. The term "computer readable medium" may represent one or more devices for storing data, including read only memory (ROM), random access memory (RAM), magnetic RAM, core memory, magnetic disk storage mediums, optical storage mediums, flash memory devices and/or other machine readable mediums for storing information. The term "computer-readable medium" includes, but is not limited to portable or fixed storage devices, optical storage devices, wireless channels and various other mediums capable of storing, containing or carrying instruction(s) and/or data.

Furthermore, embodiments may be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware or microcode, the program code or code segments to perform the necessary tasks may be stored in a computer readable medium. One or more processors may perform the necessary tasks. A code segment may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a class, or any combination of instructions, data structures, or program statements. A code segment may be coupled to another code segment or a hardware circuit by passing and/or receiving information, data, arguments, parameters, or memory contents. Information, arguments, parameters, data, etc. may be passed, forwarded, or transmitted via any suitable means including memory sharing, message passing, token passing, network transmission, etc.

Claims

Claims:

1. A computer-implemented method for controlling a complex environment, wherein the environment has plural access points for entrance and exit of people into and out of the environment, fixed infrastructure elements through which people move on journeys through the environment between the access points, and variable resources by which movement of people through and between the fixed infrastructure elements and the access points can be controlled, the method including steps of: detecting the numbers of people entering the environment and exiting the environment at each of the access points over a time period; predicting, on the basis of the detected numbers of people, the numbers of people who will enter the environment and exit the environment at each of the access points in a similar time period at a predetermined time in the future; optimising an objective function to determine a configuration of the variable resources which will enable the predicted numbers of people to enter the environment and exit the environment at each of the access points at the predetermined time; and controlling the variable resources to implement the determined configuration by the predetermined time.

2. The method of claim 1 , wherein the prediction of the numbers of people who will enter the environment and exit the environment at each of the access points over the similar time period at the predetermined time in the future is also based on stored historical data of numbers of people entering the environment and exiting the environment at each of the access points in a similar time period at an equivalent time or times in the past.

3. The method of claim 1 or 2, wherein the prediction of the numbers of people who will enter the environment and exit the environment at each of the access points over the similar time period at the predetermined time in the future is also based on external variables.

4. The method of any one of the previous claims, wherein: the prediction step also includes predicting, for each access point acting as an origin of people entering the environment over the similar time period at the predetermined time the destination access points at which those people will exit the environment, and also predicting for each destination access point of an origin access point the respective fraction of the number of people entering that origin access point who will exit at that destination access point; and in the optimisation step, the optimisation of the objective function determines a configuration of the variable resources which will enable the people entering each origin access point to travel to its respective predicted destination access points in numbers corresponding to the respective predicted fractions.

5. The method of claim 4, wherein the predictions of the destination access points and the respective fractions are based on stored historical data of destination access points and respective fractions at an equivalent time or times in the past.

6. The method of claim 4 or 5, wherein the detection step includes detecting the destination access points of people entering the environment at each origin access point over the time period; and the predictions of the destination access points and the respective fractions are based on the detected destination access points and respective fractions.

7. The method of any one of the previous claims, wherein: the detection step includes detecting people classifications, and detecting the numbers of people entering the environment and exiting the environment at each of the access points over the time period in each classification; the prediction step is performed for each detected people classification; and in the optimisation step, the optimisation of the objective function determines a configuration of the variable resource which is enabling for all the detected classifications.

8. The method of any one of the previous claims, wherein the detecting, predicting, optimising and controlling steps are repeated at intervals to update the configuration of the variable resources.

9. The method of any one of the previous claims, wherein the complex environment is a train station, a port, a hospital, an airport, a shopping mall, an office building, a theme park, a ferry or cruise ship, a campus, or a conference venue.

10. The method of any one of the previous claims, wherein the complex environment has a subset of least three of the access points which are each separated from the other access points of the subset by variable resources, whereby movement of people between the access points of the subset can be controlled.

11. A computer program comprising code which, when the code is executed on a computer, causes the computer to perform the method of any one of the previous claims.

12. A computer readable medium storing the computer program of claim 11.

13. A computer system programmed to perform the method of any one of claim 1 to 10.

14. A complex environment controlled by the computer system of claim 13, the complex environment having plural access points for entrance and exit of people into and out of the environment, fixed infrastructure elements through which people move on journeys through the environment between the access points, and variable resources by which movement of people through and between the fixed infrastructure elements and the access points can be controlled. 17

15. The complex environment according to claim 14, further having sensors for detecting the numbers of people entering the environment and exiting the environment at each of the access points over a time period.