Digital cloud-based platform and method for providing cognitive cross-collaborative access to an enclosure using authentication attribute parameters and operational conditioning tags

Abstract

Intelligent digital platform (1) for providing controlled process driven network interaction and project (41, 42,..., 4i) development, wherein automatic task assignment between units (2 / 3) with network enabled devices on a secure cloud based network (5 / 51, 52,..., 5i) takes place. Each unit has a unit or user (2 / 3) account on the digital networking platform (1) with authenticated and authorized network access (5) to the digital cross-network platform (1) and the secure cloud based network (5). Each project (41, 42,..., 4i) has an assigned relationship (1121, 1122,..., 112i) with one or more units (2 / 3) included in a data repository (112) of the digital networking platform (1), wherein each assigned relationship provides a defined relationship between one or more units (2 / 3), one or more selectable tasks (1111, 1112,..., 111i) and the project (41, 42,..., 4i). A task selector (11) includes a machine learning based matching structure (112), wherein by means of the machine learning based matching structure (112) a suitable skill authentication (1511, 1512,..., 151i) is determined to perform a specific task (111i) associated with a project (10i). Based on the determined required skill authentication (1511, 1512,..., 151i) and additional trigger criteria, at least one deployment unit (31, 32,..., 3i) is determined, the additional trigger criteria including at least a location of the operating unit (21, 22,..., 2i) submitting data (1211,..., 121i) and a location of available deployment units (31, 32,..., 3i).

Description

TECHNICAL FIELD

[0001] The present invention relates to a digital cloud-based networking system and platform that provides controlled data-driven and project-driven cross-network data access, multi-user interaction, and authenticated cross-network project collaboration among a plurality of users with network-enabled devices on a secure cloud-based network. In particular, the present invention relates to a digital platform that facilitates authenticated networking collaboration in communication with a secure, cloud-based network of the digital platform, where different categories / groups of users have an assigned relationship with tasks to be performed in process-driven networking interaction and project collaboration. The establishment of a peer review relationship provides a foundation for networking collaboration and process-driven project collaboration development. The digital platform of the present invention makes it possible to overcome known process divergences of such platforms, for example, caused by lack of standardization, drive networking complexity, and missing indicia. Moreover, web-based productivity features can provide enhancements to the user environment such that assignable users to a project can collaborate and perform accounting through the digital platform according to established relationships using the platform. In particular, the present invention relates to automation and secure access to resources, workflow, human, or project management, for example, management and / or peer review of time, human resource allocation. BACKGROUND

[0002] The field of technology research and development is changing faster than ever. To compete, it is often necessary to maintain, adapt, and exchange technology, service, and data at a level that only highly specialized technical experts can operate and maintain the number of operations within a very short reaction time, particularly in the field of complex industrial facilities or plants. Such experts are often distributed throughout the world and are difficult to find. To achieve the required scale and secure collaborative interaction, there is a great need for new, more suitable technologies for a digital platform that establishes and develops extended participant organizational boundaries and technical limitations. In fact, in recent years, digital platform technology has emerged as one of the most powerful technological manifestations of the digital revolution, which is also referred to as the "Fourth Industrial Revolution", or simply Industry 4.0. Due to the rapid advancement of the cloud, mobile phones, and analytics, as well as the cost reduction of these new technologies, digital platforms are able to create the next step in growth and technological breakthroughs. Digital platforms make it possible to bring together a development, research, and collaboration community of a large number of specialized technical personnel, service personnel, developers, and engineers. Digital platform technology enables these people to create networking collaboration and development, particularly on a large scale and efficient project-specific and task-specific collaboration and development. Moreover, this technology enables a new level of networking collaboration between companies and experts from different technical fields and industrial sectors. Thus, cloud computing platforms can provide building, deployment, and management for different types of applications and / or processes.

[0003] The increasing importance of digital platforms and associated ecosystems is based on the emergence of new digital infrastructures (e.g. Internet of Things, cloud computing, blockchain, big data analytics) and the injection and driving of digital technologies into products, services and processes. The availability of ubiquitous digital infrastructures as a basis for digital platforms fundamentally restructures the nature, methods, processes, structures and costs of providing services and conducting business. Therefore, there is a great need for proper technologies for digital platforms that allow for controlled and structured networking interactions between participants in a virtual ecosystem.

[0004] Automated service provider platform systems are known, for example, from US 2015 / 0286979 A1 or US 7,502,747 B1 ; these publications disclose the automated scheduling of service tasks, respectively the automated scheduling of appointments for service tasks. The described prior art systems are configured to manage the scheduling of jobs and the assignment of jobs to service providers based on one or more factors, for example the initial estimated length of time to perform a certain type of service task, a technician availability schedule, a technician skill or technician skill level related to a service task requested by a purchaser of a service task. However, the purchaser of a service task has to rely on data presented by the conventional service provider platform system, including all ratings and / or skill levels of service providers, wherein these ratings and / or skill levels are given based on the input of the service providers. These data relate, inter alia, to service provider data related to self-assignment of service tasks, presented service provider skill levels regarding assigned service tasks, the number of successfully completed service tasks, time requirements for performing service tasks, respective travel times and travel expenses, cost of service tasks, service provider reading / writing and reading language skills. It should be noted that this list of data is not claimed to be complete and, furthermore, one or more of the above cited data can also be of less interest. But what is common to all of the above data is that these data are provided to the purchaser without any proof or certification; therefore, the discussed data are quasi "one-size-fits-all" and will not be adjusted according to the situation based on currently generated data. SUMMARY

[0005] The present application is presented in view of the above technical problems of the prior art systems. One object of the present application is to provide a digital networking platform and a digital service provider platform for intelligent task management and technical service demand management respectively to enable project driven cross network data access, multi-user interaction and authenticated project collaboration among multiple users with network enabled devices on a secure cloud based network. In particular, the present application has the object to provide an intelligent digital platform facilitating authenticated networking collaboration in a multi-user environment through secure cloud based network communication. The present application should be able to provide collaboration in process driven network interaction and project collaboration for different heterogeneous categories / groups of users or units having an assignment relationship, in particular a hierarchical relationship, to tasks or jobs to be carried out or performed in a project.

[0006] According to the application, these objects are achieved in particular by the features of the independent claims. Further advantageous embodiments can be gathered from the dependent claims and the related description.

[0007] According to the present invention, the above mentioned objects are achieved in that a digital networking platform provides a controlled project-driven network interaction and / or shell communication, in particular by means of the present invention, a controlled project-driven network interaction between an operating unit and a deployment unit is provided by the digital platform, the units evaluate the digital networking platform via a data transmission network by means of a network-enabled device and each unit has a unit account assigned in a communication database of the digital networking platform, the unit account has associated authentication and authorization credentials for a controlled network access to a secure, cloud-based network provided by the digital cross-network platform, wherein a project saved in a persistent storage device of the digital networking platform comprises at least one assignment relationship between the operating unit, one or more deployment units and a project task, the digital networking platform comprises a network interface of the digital platform to provide a network access to a secure, cloud-based network for the operating unit and the deployment units via a data transmission network by means of a network-enabled device, wherein the project saved in the persistent storage device of the digital networking platform is only accessible for the assigned units via a dedicated secure network to provide an upload access and a download access to project data related to the project and to share the project data with other units according to the assignment relationship, a project manager module to extract project data from a project submission of the operating unit requesting a new project, wherein the new project is generated based on the project submission data and wherein the related project data is stored to the persistent storage device, a task selector to select one or more tasks, e.g. a flat screen maintenance, to be performed by the deployment units from a task database based on the project submission data and to select one or more deployment units based on an authentication skill endorsement data of an authentication skill executable by a specific deployment unit based on an indication of the deployment units, wherein the selected tasks are assigned to the selected at least one deployment unit to provide a relationship associated with the new project, and an access controller to generate a secure, cloud-based network for the new project to provide an access to the new project for the operating unit and the selected deployment units represented in said relationship, wherein a relationship between any pair of units is established before allowing any communication between said pair of units. The task selector may, for example, comprise a machine learning based matching structure, wherein by means of the machine learning based matching structure a proper skill authentication is determined to perform a specific task associated with the project and wherein at least one deployment unit is determined based on the determined required skill authentication and additional trigger criteria, the additional trigger criteria comprise at least a location of the operating unit submitting the data and a location of available deployment units. For selecting the deployment units, a mandatory skill and an optional skill may be determined by the machine-based matching structure to perform the selected task, wherein the mandatory skill is required to complete the selected task and wherein the optional skill comprises at least a reading language parameter and / or an additional skill parameter indicating an additional service or device.The deployment units can be screened by the machine-based matching structure for mandatory skills, i.e. deterministic, while a scoring structure is applied by the machine-based matching structure for optional skills, wherein the higher the score triggered for a deployment unit, the more likely the deployment unit is selected by the machine-based matching structure. The score can for example be based on best skill match. Alternatively or additionally, the score is based on highest margin match, taking into account pricing parameter values and / or cost parameter values and / or travel time or geographical distance parameter values of the deployment units. The score is also based on best fit platform growth structure to provide higher scores to deployment units that use the platform rarely or never to perform tasks. The machine-based matching structure can for example comprise one or more machine-based learning structures (ML). The machine-based learning structures (ML) can for example comprise hill climbing, genetic or evolutionary machine-based learning structures. The digital networking platform can for example comprise a dispatcher for automatically notifying the selected one or more deployment units and / or one or more service providers associated with the appropriate deployment units. The dispatcher comprises an auction process to provide access to a particular task to the appropriate deployment units.

[0008] The present invention has in particular the advantage that the digital networking platform provides a software platform to deployment units, e.g. service companies or independent skilled people, where they can be trained and competent for specific standardized (through the environment of the digital platform) service tasks / jobs. These tasks / jobs are then offered to qualified (and certified) deployment units, e.g. technicians or their service companies. Depending on the availability of these tasks / jobs, the deployment units can accept a job and then perform the task / job. The performance of the task / job is done under the management of the digital platform, e.g. wearing a uniform provided or accepted by the digital platform standards. The digital platform can for example charge the operating unit as the end customer of the service according to previously agreed rates and pay the deployment units, e.g. service technicians or companies. The responsibility for the performance of the task / job can for example be taken by the digital platform.

[0009] As a variant of the implementation, the current service provider platform comprises a service task database and a service provider interface connected to an input database designed to collect and store input data requested and / or received from at least one service provider via the service provider interface, wherein these input data are analyzed in an evaluation unit to assign at least one skill level to the service provider, wherein the service provider platform further comprises a training unit in which at least one, for example by following a predefined algorithm or configuration, or by direct input or via a feedback loop in response to the input of one or more other software components, is automatically generated or calculated with little or no human interaction, i.e. inferred, and an individually compiled training session based on the at least one skill level becomes available to the service provider automatically or on request, wherein the training session must be successfully attended by the interaction of the service provider, wherein at least another skill level is assigned to the service provider, wherein the service provider platform further comprises a matching unit designed to run at least one intelligent algorithm, wherein at least one feasible service task is assigned to the service provider in the matching unit based on the at least one and / or another skill level used in the at least one intelligent algorithm automatically or on request, and wherein the service provider platform further comprises a service provider profile database, wherein the service provider with a large amount of relevant service provider data is automatically or on request registered in the service provider profile database. The highlight of the current service provider platform is self-detection and automatic improvement. Each service provider intended to be part of the service provider platform is required to transfer its own input data into the input database, where these input data are interrogated by the evaluation unit. Subsequently, the service provider must attend a training session to "end of day" reach another skill level and based thereon, get at least one assigned service task. The new service provider platform is equipped with a self-detection unit function, respectively a self-monitoring unit function for automatically improving itself and at the same time permanently tidying up or cleaning up the entire service provider platform.

[0010] The training sessions that the service providers have to attend can be held online using a bidirectional training communication and the training sessions can include teaching material in the form of video files as well as audio files and the associated equipment can include a display, a loudspeaker, a microphone (earphone) and a mouse pointer. The training sessions can also involve lectures by using remote transmission, wherein the lecturer is an external (third party) service provider, which is hired to present an automated and individually compiled training session. Of course, also an internal lecturer can be envisaged, which can give live lectures by broadcast and / or lecture recordings for time-shifted learning for multiple students of different time zones. The service providers belonging to the service provider platform system can come from different technical fields, e.g. holding different positions; for one technical field, the service providers can act as lecturers, in another technical field, the service providers have the identity of students, while in yet another technical field, the same service provider is neither a lecturer nor a student, but only acts as a service provider performing service tasks according to the orders of the buyers.

[0011] The buyers order service tasks via the service task ordering interface and send feedback data to the service provider platform system after completion of the service task, according to which the involved service providers are re-evaluated; of particular importance is that the peer unit has a feedback connection to the evaluation unit for the automated re-evaluation of the service providers to automatically increase the service provider data and / or for the automated re-evaluation of the service providers to require the service providers to attend at least a second automated and individually compiled training session, wherein the at least second automated and individually compiled training session has to be successfully attended by the interaction of the service providers, wherein a further skill level is assigned to the service providers, wherein the further skill level automatically increases the relevant service provider data in the service provider profile database.

[0012] In another embodiment variant, the digital platform and the at least one service provider platform are structured as a distributed network using edge computing, wherein the service provider platform is implemented as a decentralized deployment to the edge servers of the providers. One advantage is that the embodiment variant can provide a low latency closer to the request or shell communication and a real-time processing of data. The digital platform operates according to big data, while the edge computing structure operates according to "immediate data", which are real-time data generated by sensors associated with the operating units and / or deployment units of the evaluation digital networking platform. For example, sensors or measuring devices can capture geographical position parameters like longitude or latitude measurement parameters or operating time parameters of one or more operating units, etc. Based on the additional measurement parameters, the digital platform can provide additional selectivity regarding the provided or possibly optional shell communication, i.e. secure, cloud-based network, between the operating units and / or deployment units with respect to a particular project.

[0013] Further, the digital platform can provide shell icon communication, e.g. via a remote sync client at the operating unit and / or deployment unit, which involves a status or other parameters of a project hosted on the cloud-based digital platform. The respective host server application at the cloud-based collaborative digital platform can utilize the project provided via the digital platform to gradually update the remote sync client at the operating unit and / or deployment unit (or a device running in association with the operating unit and / or deployment unit). The digital platform, including the cloud-based collaboration and cloud storage server, can utilize events occurring via the digital platform to gradually update the remote sync client and / or access tool at the operating unit and / or deployment unit. The operating unit and / or deployment unit can include any digital, network-enabled system and / or device, and / or any combination thereof to establish a communication or connection to the digital platform, including a wired connection, a wireless connection, a cellular connection with another device, a server and / or other system, e.g. a host server and / or a notification server application on the digital platform, and / or a measurement device and sensor associated with the operating unit and / or deployment unit. A received or detected signal can e.g. indicate an activity of the operating unit and / or deployment unit to a project collaborator accessing the project via the digital platform providing a web-based collaboration environment or online collaboration platform. Further, the digital platform and / or the operating unit and / or deployment unit and its database system can be vulnerable to an intrusion by an unauthorized user or system, e.g. by a reverse shell connection, enabling an intruder to execute low-level commands on the digital platform and / or the operating unit and / or deployment unit. A reverse shell connection can be detected by monitoring and inspecting packet data traffic between a network of the digital platform and an external network. Such a process can e.g. detect a reverse shell connection exploiting an initial shell detection by analyzing the transmission direction and payload size of the monitored packet sequence with respect to a predetermined traffic pattern after detecting a normal shell session originating from inside the network of the digital platform. Particular patterns can be selected for the different systems involved, e.g. the digital platform and / or the operating unit and / or deployment unit. BRIEF DESCRIPTION OF DRAWINGS

[0014] The application will be explained in more detail below by means of examples and with reference to the drawings, in which:

[0015] Figure 1 and Figure 2 Each shows a diagram schematically illustrating a dynamic task allocation and / or arrangement of submitted projects 101, 102,..., 10i by means of a digital networking platform 1, wherein the focus is on level 1 tasks 1111, 1112,..., 111i, wherein the tasks 1111, 1112,..., 111i are grouped into two or more (here three) levels of tasks 1111, 1112,..., 111i.

[0016] Figure 3 The following diagram is shown, which schematically illustrates the authentication process for the deployment units 31, 32,..., 3i.

[0017] Figure 4 The following diagram is shown, which schematically illustrates the dynamic task allocation process requested by means of the digital networking platform 1 for submitted projects 101, 102,..., 10i relating to tasks 1111, 1112,..., 111i.

[0018] Figure 5 The following block diagram is shown, which schematically illustrates the digital networking platform 1 for providing a controlled flow-driven networking interaction between units 2 / 3 having network-enabled devices accessing each project 101, 102,..., 10i through a network 4, said projects 101, 102,..., 10i having an allocation relationship 1011, 1012,..., 102i with one or more units 2 / 3 included in the persistent storage device 10 of the digital networking platform 1, each allocation relationship providing a defined relationship 1011, 1012,..., 102i between at least two units 2 / 3 and a project 101, 102,..., 10i.

[0019] Figure 6 The following block diagram is shown, which schematically illustrates the digital networking platform 1 implemented as a digital service provider platform 1, which comprises a service task database 1.1 and a service provider interface 1.2 connected to an input database 1.3 designed to collect and store input data 1.31, 1.32, 1.33 requested and / or received from at least one service provider 1.21, 1.22, 1.23 through the service provider interface 1.2, wherein these input data 1.31, 1.32, 1.33 are analyzed in an evaluation unit 2.1 to assign at least one skill level 2.11, 2.12, 2.13 to the service provider 1.21, 1.22, 1.23.

[0020] Figure 7The following figure is shown which schematically illustrates the digital networking platform 1 for providing a controlled, flow-driven networking interaction and project 101, 102,..., 10i development between the operating unit 21 and the deployment units 31, 32,..., 3i with network-enabled devices 2i2 / 3i2 on the secure cloud-based network 51 and 5. Each unit 2i / 3i has a unit account 162 / 163 in the digital networking platform 1 with assigned authentication and authorization credentials for a controlled network access 141 to the digital cross-network platform 1 and the secure cloud-based network 5 for authentication and authorization. Each project 101, 102,..., 10i has assigned relationships 1121, 1122,..., 112i with one or more units 2 / 3 included in the persistent storage 10 of the digital networking platform 1, wherein each assigned relationship 1121, 1122,..., 112i provides a defined relationship between the operating unit 21, the one or more deployment units 31, 32,..., 3i and the project 101, 102,..., 10i.

[0021] Figure 8 and Figure 9 The following figure is shown which schematically illustrates the digital networking platform 1 generating and storing a digital replica of the project 101, 102,..., 10i, wherein structural, operational and / or environmental state parameters 10ii of the real-world project 101, 102,..., 10i are measured, monitored and transmitted to the digital platform 1 by means of at least one sensor associated with the twin physical replica of the project 101, 102,..., 10i, wherein the digital replica is dynamically updated based on the transmitted parameter values and wherein a predictive time series of parameter values for future development of the project 101, 102,..., 10i is generated.

[0022] The following description and drawings are illustrative and are not to be construed as limiting. Numerous specific details are described to provide a thorough understanding of the present disclosure. However, in certain instances, well known or conventional details are not described in order to avoid obscuring the description. References to one or an implementation or embodiment in the present disclosure can be, but not necessarily, references to the same implementation or embodiment; and, such references mean at least one of the implementations. Reference to “one implementation” or “an implementation” in the specification means that a particular feature, structure, or characteristic described in connection with the implementation is included in at least one implementation of the disclosure. The appearances of the phrase “in one implementation” in various places in the specification are not necessarily all referring to the same implementation, nor are they necessarily all referring to a particular implementation or embodiments that are mutually exclusive or alternative implementations. Furthermore, various features might be described which can be exhibited by some implementations and not by others. Similarly, various requirements might be described which can be requirements for some implementations but not for others. The terminology used by the inventor in this specification were chosen for reasons of readability and DETAILED DESCRIPTION

[0023] Figures 1 to 7The architecture of possible implementations for embodiments of a digital networking platform 1 for providing a controlled project-driven networking interaction between operating units 2 and deployment units 3 is schematically shown. Structurally, the present invention extends the prior art secure collaboration network within a new technical solution to include different levels of units 31, 32,..., 3i with different skills for conducting specialized tasks 1111, 1112,..., 111i with different networks, different organizations, different machines, and with different geographical locations and other boundary condition contexts. The present invention thus enables units / users 2 / 3 to securely exchange and distribute confidential information, generate and submit project development processes with assigned tasks to be performed by deployment units 3, etc. across company boundaries and geographical boundaries through a secure connection of peer-reviewed tasks and corresponding certified deployment units. Operating units 2 and deployment units 3 assess the digital networking platform 1 via a data transmission network 4 through network-enabled devices 2i2 / 3i2.

[0024] As a possible implementation of the physical hardware of the digital platform 1, the digital platform 1 comprises a web server 17 that can be connected to the data transmission network 4 via a secure data transmission network interface 18 and a firewall 171. The firewall 171 is connected to the data transmission network 4, in particular the global backbone network Internet 41, via a router 172. The authorized units 2 / 3 can submit and / or modify and / or access projects 101, 102,..., 10i stored in a persistent storage 10 as a project database, which can be accessed via the web server 17 through a secure connection on the data transmission network 4. The persistent storage 10 can be used to hold a distributed, e.g. international, networked collaboration of more than one project 101, 102,..., 10i, wherein a secure collaboration environment within a project company is provided by means of the digital platform 1 over the backbone Internet 41. The authorized units 2 / 3 are units that have a registered and authorized account on the digital platform 1. The authorized units 2 / 3 can access the digital platform 1, e.g. using web-enabled devices, including desktop or laptop computers, personal digital assistants (e.g. iPad, etc.), web-enabled cellular phones and / or digital phones, and other web-enabled devices or data transmission network 4 enabled devices. Herein, a "web-enabled device" means a device that is capable of browsing the Internet using an Internet browser, while a "network-enabled device" generally refers to a device that allows access to the digital platform 1 over the data transmission network 4. Furthermore, the digital platform 1 can comprise automated productivity tools that are modules that supplement the unit 2 / 3 account 162 / 163 with professional enhancements for facilitating interactions between different users 2 / 3 of the digital platform 1, e.g. interactions between operating units 2 and deploying units 3, etc. In embodiments, the web server of the digital platform 1 is implemented as a secure server. For this purpose, a "secure server" can for example mean a server that is registered with a digital certificate authority for the purpose of being an authentication server and providing secure transactions over the Internet. Secure data transmission over the network 4 can also include appropriate encryption / decryption procedures to ensure secure networks 51, 52,..., 5i. In summary, the web interface of the digital cross-network platform 1 can for example be a web interface and the network-enabled device can be a web-enabled device, wherein the digital networking platform 1 comprises selectable productivity tools for interfacing with the projects 101, 102,..., 10i. In addition, the productivity tools that are accessible from the web interface can also include a task manager module 143 and a collaboration module 144 and a file management module 145, as described herein.

[0025] Each unit 2 / 3 has a unit or user account 162 / 163i on the digital networking platform 1, which has assigned authentication and authorization credentials for controlled network access 5 to the digital cross-network platform 1 and the secure cloud-based network 5 for authentication and authorization. The authentication and authorization credentials can for example comprise at least a username and password in a secure database. However, other authentication and authorization variants, for example biometric authentication and authorization, are also conceivable. In particular, the network-enabled device 2i2 / 3i2 and the digital platform can be implemented as a network node of the data transmission network 4, which comprises a physical network interface 2i21 / 3i21 / 18, for example to a LAN, WLAN (Wireless Local Area Network), Bluetooth, GSM (Global System for Mobile Communications), GPRS (General Packet Radio Service), USSD (Unstructured Supplementary Service Data), UMTS (Universal Mobile Telecommunications System) and / or Ethernet or another wired LAN (Local Area Network) or the like. The reference 4 can be based on IEEE 802.11 or another standard or can comprise different heterogeneous networks, such as for example a Bluetooth network for example for devices within the area of a roof covering, a mobile radio network with GSM and / or UMTS and / or LTE (Long Term Evolution; also referred to as 3.9G) or the like, a wireless LAN for example based on IEEE wireless 802.1x, or in addition a wired LAN, i.e. a local fixed network, in particular also a PSTN (Public Switched Telephone Network) or the like. The interface 2i21 / 3i21 / 18 of the network node 1 / 2 / 3 can be not only a packet-switched interface, for example those interfaces used directly by a network protocol such as Ethernet or Token Ring, but also a circuit-switched interface that can be used with a protocol such as PPP (Point-to-Point Protocol, see IETF RFC), SLIP (Serial Line Internet Protocol), GPRS (General Packet Radio Service), TCP / IP (Transmission Control Protocol / Internet Protocol), i.e. for example those interfaces that do not have a network address such as a MAC address or a DLC address. As mentioned in part before, the communication can take place for example through a LAN; for example also by means of special short messages such as SMS (Short Message Service), EMS (Enhanced Message Service); through a signaling channel such as USSD (Unstructured Supplementary Service Data) or other technologies like MExE (Mobile Execution Environment), GPRS (General Packet Radio Service), WAP (Wireless Application Protocol) or UMTS (Universal Mobile Telecommunications System); or through IEEE wireless 802.1x or via another digital information channel.

[0026] Each unit 2 / 3 has a unit account 1621 / 1631 assigned in the communication database 161 of the digital networking platform 1 with associated authentication and authorization credentials provided by the digital cross-network platform 1 for controlled network access 5i1 to the secure, cloud-based network 5 / 51, 52,..., 5i. Items 101, 102,..., 10i saved in the persistent storage 10 of the digital networking platform 1 include a specific operating unit 2i, one or more deployment units 3i and at least one assignment relationship 10i1 between the item 101, 102,..., 10i and the deployment unit 3i.

[0027] The digital networking platform 1 comprises a network interface 18 in the digital platform 1 to provide network access to the secure, cloud-based network 5 / 51, 52,..., 5i via data transmission network 4 for operating units 21, 22,..., 2i and deployment units 31, 32,..., 3i by means of network-enabling devices 2i2 / 3i2. Items 101, 102,..., 10i saved in the persistent storage 10 of the data networking platform 1 are only accessible for the assigned units 2 / 3 via the dedicated secure network 51, 52,..., 5i to provide upload access and download access for item data 10ii related to the item 101, 102,..., 10i and to share item data 10ii with other units 2 / 3 according to the assignment relationship 10i1.

[0028] The digital networking platform 1 comprises a project manager module 14 to extract project data 10ii from a project submission 10i2 of an operating unit 2 requesting a new project 101, 102,..., 10i. The new project 101, 102,..., 10i is generated by a project analyzer 142 in the project manager module 14 based on the project submission data 10i2 and wherein the relevant project data 10ii is stored to the persistent storage 10. The submission data 10i2 can be submitted to the digital networking platform 1 from the operating unit 2i,..., 21, for example by means of the operating unit access interface 12, wherein the corresponding project 101, 102,..., 10i is generated by means of a submission analyzer module 122 and assigned to the persistent storage 10. The relationship between the units 2 / 3 with respect to the projects 101, 102,..., 10i is defined on an individual basis and based on the authentication characteristics assigned to a specific deployment unit 3, so that each pair of units 2 / 3 has a defined relationship. Each relationship establishes a hierarchical or equal relationship between two users, for example a deployment unit 3, an authenticated technician, engineer or maintenance personnel for example for a specific task, an operating unit 2, etc. By structuring the digital platform 1 according to the unit 2 / 3 relationship 10ii, essentially a secure network structure 5 is given by the relationship structure. Instead of a single network with many users, the digital platform 1 defines multiple user-centric networks 5 / 51, 52,..., 5i, which are similar to the real-world network 5 of units 2 / 3 as appropriate network nodes. For authentication, the digital networking platform 1 can for example comprise an authentication register 15 with a task authentication data repository 151 holding for each task 1111, 1112,..., 111i in the task database 111 a deployment unit 3 accessible task authentication 1511, 1512,..., 151i and an authentication evaluation module 152 comprising for each task 1111, 1112,..., 111i in the task database 111 an authentication evaluation process for approving the assignment of a task authentication 1511, 1512,..., 151i for a deployment unit 31, 32,..., 3i. Upon approval of a task authentication 1511, 1512,..., 151i, a respective authentication task approval 1632 is assigned to the deployment unit 31, 32,..., 3i for a specific task.

[0029] The digital networking platform 1 comprises a task selector 11 selecting one or more tasks 111i from a task database 111 to be performed by a deployment unit 3 based on project submission data 1012and a task manager module 143 selecting one or more deployment units 3 based on authentication skill approval data 1632of authentication skills 31i1that can be performed by a specific deployment unit 3 based on an indication of the deployment unit 3. The selected tasks 111i are assigned to the selected at least one deployment unit 3 providing a relationship 1011associated with the new project 101, 102,..., 101. The task selector 11 can for example comprise a machine-based matching structure 112, wherein a suitable skill authentication 1511, 1512,..., 151i is determined to perform a specific task 111i associated with the project 101 by means of the machine-based matching structure 112. The at least one deployment unit 31, 32,..., 31may be determined for example based on the determined skill authentication 1511, 1512,..., 151i needed and additional trigger criteria comprising at least a location of the operating unit 21, 22,..., 2i of the submission data 1211,..., 121i and a location of the available deployment units 31, 32,..., 31. As a variant of implementation, for selecting the deployment units 31, 32,..., 31, a mandatory skill and an optional skill 1511, 1512,..., 151i can be determined by the machine-based matching structure 112 to perform the selected task 111i. The mandatory skill 1511, 1512,..., 151i is necessary to complete the selected task 111i. The optional skill 1511, 1512,..., 151imay comprise at least a reading language parameter and / or an additional skill parameter indicating an additional service or device. The deployment units 31, 32,..., 31may be screened for example by the machine-based matching structure and matcher 112 for the mandatory skill 1511, 1512,..., 151i. In contrast, a scoring structure can be applied by the machine-based matching structure 112 for the optional skill 1511, 1512,..., 151i, wherein the higher the score triggered for a deployment unit 31, 32,..., 31, the more likely the deployment unit 31, 32,..., 31is selected by the machine-based matching structure 112. The score may, for example, be based on a best skill match. Alternatively or additionally, the score may, for example, be based on a highest margin match taking into account pricing parameter values and / or cost parameter values and / or travel time or geographical distance parameter values of the deployment units 31, 32,..., 31. Additionally or alternatively, the score may, for example, also be based on a best fit platform growth structure to provide a higher score to a deployment unit 31, 32,..., 31that uses the platform 1 to perform tasks 111i rarely or never.The machine-based matching structure 112 can for example comprise one or more artificial intelligence-based learning structures (AI), in particular machine-based learning structures (ML). The machine-based learning structures (ML) can for example comprise hill climbing, genetic or evolutionary machine-based learning structures or other AI-based structures. Once the operating unit / customer 2 orders a service, submits a project request, the custom business logic of the digital platform finds out the skills necessary to complete that service job. These skills are fed to the matcher structure of the digital platform 1 together with other data like the location of the operating unit / customer 2 and the location of the deployment unit / technician 3. Then, these required skills are divided into two groups: mandatory skills and optional skills. Typically, mandatory skills are required to complete the actual job in high quality, while optional skills are typically additional skills to speak the customer language fluently using the operating unit 2, the operating unit / customer 2 has but did not order. Deployment units / technicians 3 are filtered out that do not fulfill the mandatory skills. For the other aspects, the described scoring system is used. The higher the score, the more relevant the technician is for the job or the more suitable the technician is for the job. Optional skills are seen as “additional points” or “bonus points” in the scoring system. The matcher 112 can use different scoring algorithms to produce different rankings. After the “best technician for the job” matching (best skill matching), there is also a “highest margin” scoring, where the pricing of the technicians is considered as well as the fees and travel times. The “best for platform growth” algorithm gives higher scores to technicians that can have theoretical expertise but never actually performed jobs in the field. This algorithm will create a balanced mix of experienced technicians and beginners to ensure on-the-job education (“learning by doing”) without sacrificing any level of quality. These individual algorithms can be implemented as mathematical optimization algorithms like hill climbing, genetic or evolutionary algorithms or machine learning (ML) based. The result is one or more ranked technicians with jobs suitable for them. This result is then passed to the dispatcher 19.

[0030] As a variant, the matcher 112 comprises a new two-layer matching structure. In a first step, the matching parameters of the operating units 2, such as geolocation measurement parameters, such as longitude and latitude parameters provided for example by a GSM module, skill specification parameters and / or skill rate parameters, user-specific parameters (age, mobility, etc.) associated with the operating units 2, etc., can be clustered in the context of the assigned project-specific parameters, for example using an unsupervised machine learning structure. The clusters can then be classified by an expert system based on historical data, or can be classified manually by a human expert. In a second step, the matcher comprises a supervised learning structure to allow a predictive generation of project development parameters, such as for example indicating a predicted time development under the respective automation job or operation schedule, or a probability of occurrence of external events affecting the development of the project, in particular the time development.

[0031] The matcher 112 can comprise and access dedicated measurement devices and sensors to monitor the progress of the projects 101, 102,..., 10i by measuring appropriate measurement parameters. The matcher 112 can be implemented to comprise one or more digital analysis structures for the analysis and quality control of the development of the projects 101, 102,..., 10i and thus for example to ensure an optimized project result during processing. Such monitoring can for example include triggering measurement parameters to comply with predetermined project objectives, boundary condition values or rules. The monitoring and analysis measurement means used can include a wide range of technologies, in particular optical measurements (e.g. camera, i.e. optical sensor, light source, etc.), chemical and mechanical and engineering dynamic adaptation and improvement indicated by appropriate project 101, 102,..., 10i parameter patterns or projects 101, 102,..., 10i and / or operation unit 2 parameter pattern recognition for example by machine-based learning (ML), etc. In addition to AI (artificial intelligence) based methods, other principles for real-time recognition of acquired parameter values and / or measurement data are available.

[0032] With ML-based measurement technology, the matcher 112 can for example use machine learning to recognize patterns and regularities based on captured and / or measured project 101, 102,..., 10i and / or operating unit 2 parameter values. Suitable ML-based systems can also evaluate unknown data. The matcher 112 can also use techniques such as "knowledge discovery in databases" and "data mining" to find new patterns and / or anomalies in the development of the projects 101, 102,..., 10i, which are in particular related to finding new patterns and regularities. As a variant, the matcher 112 uses the method of "knowledge discovery in databases" to generate or pre-process learning data for "machine learning". In turn, techniques from machine learning can be applied to data mining later, for example.

[0033] In another variant, the matcher 112 comprises a deep learning based structure which, for the present invention, is a possible learning variant of the ML based device using artificial neural networks. Deep learning is also known as multi-layer learning or deep learning, which is a machine learning technique using artificial neural networks (ANN) having many hidden layers between the input layer and the output layer, which can form a specific internal structure capable of storing knowledge adaptively in the learning process. With the techniques discussed above, the parameter value patterns and / or anomalies in the development of the projects 101, 102,..., 10i can be filtered out and avoided or at least detected and generate corresponding warning signaling of the development defects of the projects 101, 102,..., 10i. As previously mentioned, the present invention not only relates to enabling secure cross-network collaboration and intelligent matching between the projects 101, 102,..., 10i and / or the operating units 2 and / or the deployment units 4, but also particularly relates to the measurement and analysis techniques for the analysis, monitoring and automated quality control of the projects 101, 102,..., 10i and real-time development of the projects, as well as the predictive generation of the project development parameters, enabling the predictive measurement of the probability of the occurrence of risk events affecting the development of the projects based on the measured parameters measured at the projects 101, 102,..., 10i.

[0034] The digital networking platform 1 comprises an access controller 141 generating a secure, cloud-based network 5 / 51, 52,..., 5i for the new project 101, 102,..., 10i to provide access to the new project 101, 102,..., 10i for the operating unit 2 and the selected deployment unit 3 represented by the relationship 10i1, wherein the relationship 10i1 is established between the pair of units 2 / 3 before any communication between the pair of units 2 / 3 is allowed.

[0035] As a further variant, the digital networking platform 1 can for example comprise a dispatcher 19 for notifying the selected one or more deployment units 31, 32,..., 3i and / or the service provider associated with the appropriate deployment unit 31, 32,..., 3i. The dispatcher 19 can for example comprise an auction process which offers access to a particular task 1 1 1 i to the appropriate deployment unit 31, 32,..., 3i. Since it cannot be guaranteed that the deployment unit / technician 3 with the highest score is actually available for a job, the dispatcher 19 has to efficiently find a deployment unit / technician 3 with the highest score or a sufficiently high score which is actually available for performing the task / job 1 1 1 1, 1 1 1 2,..., 1 1 1 i and which accepts the task / job 1 1 1 1, 1 1 1 2,..., 1 1 1 i. This is combined with legal constraints in certain countries, for example Germany, in which such a digital platform is not allowed to directly or in certain cases even indirectly address any selected deployment unit / technician 3, for example the dispatcher component 19 contacts the planner of a service company to coordinate the job acceptance. In this case, the planner of the service company technically acts as a deployment unit 3 according to the present application. Different algorithms can for example be used to efficiently perform such a process, for example an auction-like process, in which a task / job 1 1 1 1, 1 1 1 2,..., 1 1 1 i is offered at a certain price and the fastest service company which is a deployment unit 3 accepting this job starts to perform this job. Other algorithms can for example optimize different aspects of the service network of the digital platform, for example offering a task / job 1 1 1 1, 1 1 1 2,..., 1 1 1 i first to new members or members with a high rating of the platform 1 or spreading the task / job 1 1 1 1, 1 1 1 2,..., 1 1 1 i among the participating members, i.e. deployment units 3, to maximize the fairness of the allocation.

[0036] As previously mentioned, the assignment relationship 10i1may for example define a hierarchy comprising at least two subgroups 2 / 3, wherein the first subgroup 2 comprises a plurality of operating units 21, 22,..., 2i defined by the unit account 162i and the second subgroup 3 comprises a plurality of deployment units 31, 32,..., 3i defined by the unit account 163i. The secure cloud-based network access 5 / 51, 52,..., 5i provided via the data transmission network interface 18 for a specific project 101, 102,..., 10i can for example be different for the operating units 21, 22,..., 2i of the first subgroup 2 and the deployment units 31, 32,..., 3i of the second subgroup 3. The submission data 10i2may for example be transmitted by the operating unit 2 to the digital networking platform 1 by means of the operating unit access interface 12, wherein the respective project 101, 102,..., 10i is generated by means of the project analyzer 142 and assigned to the persistent storage 10. One or more tasks 1111, 1112,..., 111i of the task database 111 can for example be selected by the task selector 11 based on the project data 10ii associated with the project 101, 102,..., 10i to be executed by the deployment unit 31, 32,..., 3i. At least one deployment unit 31, 32,..., 3i can for example be selected from the deployment unit profile database 163 based on the deployment unit profile 1631, 1632,..., 163i to execute the selected one or more tasks 1111, 1112,..., 111i, wherein the deployment unit profile 1631, 1632,..., 163i matches the selected one or more tasks 1111, 1112,..., 111i and generates the related relationship data 10i1 defining the relationship 10i1 between the new project 101, 102,..., 10i and the tasks 1111, 1112,..., 111i to be executed. To select at least one deployment unit 31, 32,..., 3i from the deployment unit profile database 163 according to the deployment unit profile 1631, 1632,..., 163i, each deployment unit profile 1631, 1632,..., 163i can for example comprise at least one authentication task approval 1632 for one selectable task 1111, 1112,..., 111i of the task database 111.

[0037] The digital platform 1 can further comprise, for example, a quote module 146 or a quote server interface 1461 for periodically interacting with a quote module or a quote server, for retrieving financial information from the quote module or the quote server, and for storing the retrieved financial information in a persistent storage assigned to the project 101, 102,..., 10i. The account module 16 can further comprise, for example, means for displaying and accounting financial information including financial account information related to the unit 2 / 3 or the user account 162 / 163 and a cost base, and valuation data related to the financial account information generated from the financial account information using the retrieved task characteristic data associated with the specific project 101, 102,..., 10i.

[0038] To predictively generate future states of the projects 101, 102,..., 10i based on captured or measured project parameters, the digital platform 1 and the matcher 12 can further comprise a digital twin of each project 101, 102,..., 10i, respectively. To this end, by means of the digital platform 1, the digital twin or digital twin representation of the project 101, 102,..., 10i is analyzed to provide a measure for the future state or operation of the twin real-world project 101, 102,..., 10i based on a value time series generated from values within a future time period of said future state, e.g. in relation to a probability of occurrence of predefined events for a possible development pattern of the project 101, 102,..., 10i. According to some embodiments, the digital twin of the twin physical project 101, 102,..., 10i has access to the persistent storage 10 and utilizes the probabilistic structure creation unit to automatically create a predictive structure which can be used by the digital twin modeling process to create predictive risk / occurrence probability measures. To account for resulting influences captured and measured by physical measurement parameters of measurement sensors and devices associated with the project 101, 102,..., 10i to provide project parameters on the operation and / or structure and / or environment of the twin real-world project 101, 102,..., 10i for a future time period, the cumulative predictive parameter modeling by the machine learning module can further comprise a step of detecting first anomalies or significant influences within the generated and measured time series of parameter values, wherein the detection of anomalies and significant events is triggered by a deviation of the measurement of a single or a group of operation and / or environment project parameters exceeding a defined threshold. The digital platform 1 and the matcher 112 can further detect second anomalies and significant influence events or project development parameters based on time series defining the operational state of the digital twin. By means of dynamic time warping, the topological distance between measurement time series of parameters over a period of time is determined as a distance matrix. Dynamic time warping can be implemented, e.g., based on dynamic time packing. The time series signal of measured project development steps can be matched, e.g., as a spectrum or cepstrum value tuple with other value tuples of the time series signal of measured project development steps or project influence event occurrences. The value tuples can be supplemented, e.g., with further measurement parameters, such as one or more of the current digital twin parameters and / or environment parameters discussed above. Using a weighting of the individual parameters for each measured value tuple, a difference measure between any two values of the two signals is established, e.g., a normalized Euclidean distance or Mahalanobis distance. The matcher 112 searches for the most favorable path from the beginning to the end of the two signals via the cross-distance matrix of pairwise distances of all points of the two signals. This can be done, e.g., dynamically efficiently. The actual path, i.e. the packing, is generated by backtracking after the first pass of dynamic time warping. For a mere determination, i.e. a respective template selection, a simple pass without backtracking is sufficient.However, the backtracking enables each point of one signal to be precisely mapped to one or more points of the respective other signal and thus exhibits an approximate time distortion. It should be added that in the current case, due to the algorithmic reasons of extracting the signal parameters of the value tuples, the best path through the signal difference matrix can not necessarily correspond to the actual time distortion. With the aid of the statistical data mining unit of the matcher 112, the measured and dynamically time-normalized time series are then clustered into disjoint clusters (cluster analysis) on the basis of the measured distance matrix, whereby the measured time series of a first cluster indicate virtual twin operation and states within a standard range and the measured time series of a second cluster indicate virtual twin operation and states outside the standard range. The clustering, i.e. the cluster analysis, can thus be used to assign a similarity structure in the measured time series, whereby the groups of similar measured time series found in this way are referred to here as clusters and the group assignment is referred to as clustering. The clustering with the aid of the matcher 112 is accomplished here with the aid of data mining, wherein new cluster regions can also be found by using data mining. The automation of the statistical data mining unit for clustering of the distance matrix can be implemented, for example, on the basis of a density-based spatial cluster analysis processing in the presence of noise, in particular, a density-based spatial cluster analysis in the presence of noise can be implemented on the basis of DBScan. DBScan operates on the basis of density as a spatial cluster analysis in the presence of noise and is able to detect multiple clusters. Noise points are ignored and returned individually.

[0039] As a preprocessing step, e.g. pre-processing, a dimensionality reduction of the time series can be performed. Generally, if needed by the dynamics of the twin items 101, 102,..., 10i, the analysis data described above consists of a large number of different time series, e.g. with a predefined sampling rate. Here, each variable can be divided into two types of time series, e.g.: (1) time-sliced time series, when the processing or dynamics of the twin items 101, 102,..., 10i ends, the time series can be naturally divided into smaller pieces (e.g. operating periods, daily time periods, etc.); and (2) continuous time series: when the time series cannot be divided in an obvious manner and the time series has to be processed (e.g. sliding windows, arbitrary division,...). Additionally, the time series can also be univariate or multivariate: (1) univariate time series: the observed process consists of only one observable observation sequence (e.g. a structural parameter of the twin items 101, 102,..., 10i); (2) multivariate time series: the observed process consists of two or more observable observation sequences that can be correlated (e.g. a structural parameter and a condition / state of the twin items 101, 102,..., 10i or an element of the twin items 101, 102,..., 10i).

[0040] The use of time series for the processing steps of projects 101, 102,..., 10i poses a technical challenge, especially in case the time series have different lengths (e.g. operational parameter / environmental measurement parameter time series). Therefore, in the context of the digital platform 1 of the present invention, it can be technically advantageous to use a pre-processing to pre-process these time series into a more directly usable technical format. Using a dimensionality reduction method, a latent space can be derived from a set of time series. This latent space can be implemented as a multi-dimensional space containing features encoding meaningful or technically relevant properties of the high-dimensional data set. The technical application of this idea can be found in natural language processing (NLP) methods, where a word embedding space is created from text data or in the current case a time series embedding space or in image processing, where a convolutional neural network encodes higher order features of an image (edges, colors,...) in its final layers. According to the present invention, this can be technically achieved by creating a latent space from several time series from the replay data and using this latent space as a basis for subsequent tasks like event detection, classification or regression tasks. In the current case, a latent space can be generated for time series signals using technical methods like principal component analysis and dynamic time warping as well as deep learning based technical methods like autoencoders and recurrent neural networks similar to the technical methods used for computer vision and NLP tasks.

[0041] With regard to the generation of a time series embedding space, the basic technical problem in the current case that complicates technical modeling and other learning problems is the dimensionality. The time series or sequences for which the model structure is to be tested can differ from any of the time series sequences seen during training. Technically, a possible approach can be based on n-grams that generalize by concatenating very short overlapping sequences seen in the training set together. However, in the current case, the dimensionality problem is solved by learning distributed representations of the vocabulary, which enables each training set to inform the model about an exponential number of semantically adjacent sentences. Modeling learns simultaneously: (1) a distributed representation of each time series; and (2) a likelihood function for sequences of time series represented in these representations. Generalization is achieved by assigning high probabilities to a series of time series that has never been seen before, consisting of time series similar (in the sense of approximate expression) to the ones forming the set that has been seen. Training such a large model (with millions of parameters) in a reasonable time is a technical challenge in itself. As a solution to the current case, a neural network is used, which can be used for the likelihood function, for example. With regard to two sets of time series, it can be shown that the approach used here provides significantly better results compared to n-gram models of the prior art and the proposed approach enables the use of longer time series and time series context.

[0042] In the current case, the ability of multi-layered backpropagation networks to learn complex, high-dimensional, non-linear mappings from large sets of examples makes these neural networks, in particular convolutional neural networks, a technical candidate for time series recognition tasks. However, for the application in the present invention there is the technical problem that in technical architectures for pattern recognition, usually a hand-designed feature extractor collects relevant information from the input and eliminates irrelevant variability. A trainable classifier then classifies the resulting feature vectors (or strings) into classes. In this scheme, a standard, fully connected multi-layered network can be used as a classifier. A potentially more interesting scheme is to eliminate the feature extractor, feed the "raw" input (e.g. normalized images) to the grid and rely on backpropagation to turn the first few layers into a suitable feature extractor. While this can be done with ordinary fully connected feedforward networks and has some success for the task of detecting time series, in the current context there is a technical problem. First, the measurement parameters of a time series can be very large. Thus, a fully connected first layer with a few hundred hidden units already requires several ten thousand weights. If not enough training data is available, overfitting problems arise. Furthermore, the technical requirements for storage media grow drastically with such numbers. However, the technical surface problem is that these networks have no intrinsic invariance with respect to local shifts in the input time series. That is, the preprocessing discussed above with appropriate normalization or other temporal normalization has to normalize and center the time series. On the other hand, technically, no preprocessing is perfect.

[0043] Second, the technical problem of fully connected networks is that the topology of the input time series is completely ignored. The input time series can be applied to the network in any order without affecting the training. However, in the current case, the process flow has a strong local two-dimensional structure and the measurement parameters of a time series have a strong one-dimensional structure, i.e. the measurement parameters adjacent in time are highly correlated. The local correlation is the reason why in the context of the present invention local features of a time series are extracted and combined before recognizing spatial or temporal objects. Convolutional neural networks thereby implement the extraction of local features by restricting the receptive field of a hidden unit to a local cell. In the current case, the use of convolutional networks in the recognition of time series technically ensures the realization of invariance to shifts and losses, i.e. by local receptive fields, joint weights (or weight replication) and the application of temporal subsampling of a time series. The input layer of the network thereby receives time series that are approximately time-normalized and centered (see time packing above).

[0044] As mentioned before, in order to generate a latent space for time series signals, one can choose e.g. principal component analysis and dynamic time warping or a deep learning based technique approach, e.g. using recurrent neural networks. However, in the present invention it should be noted that using recurrent backpropagation to learn information over long time intervals can take a long time in general due to insufficient decay of error feedback. Therefore, in the context of the present invention a new, efficient and gradient based approach is used. Here, the gradient is truncated at its point of no return, such that the network can learn to bridge minimal time delays of more than 1000 discrete time steps by means of implementing a constant error flow by constantly rotating the error within a certain cell. The multiplication gate cell learns to open and close access to the constant error flow from this. By this embodiment according to the present invention, the network preserves locality in space and time with respect to learning time series.

[0045] With respect to the autoencoder embodiment comprised by the matcher 112, the network is trained in an unsupervised manner (unsupervised learning) such that the input signal can first be converted to a low-dimensional latent space and reconstructed with minimal information loss by the decoder. This approach can be used to convert high-dimensional time series to low-dimensional time series by training a multi-layer neural network with a small central layer to reconstruct the high-dimensional input vector. Gradient descent can be used to fine-tune the weights of such an "autoencoder" network. However, this works well only if the initial weights are close to a suitable solution. In learning time series, the embodiment described here provides an effective method of initializing the weights that allows the autoencoder network to learn a lower-dimensional code that performs better than principal component analysis as a tool for reducing the dimensionality of the data. The dimensionality reduction of time series according to the present invention facilitates classification, visualization, communication and storage of high-dimensional time series. One possible approach is principal component analysis (PCA), which finds the directions in which the variation in the time series is largest and represents each data point by its coordinates along each of these directions. As an embodiment variant, one can use a nonlinear generalization of PCA by using an adaptive multi-layer "encoder" network to convert the high-dimensional time series to a low-dimensional code and a similar decoder network to recover the time series from the code. In the embodiment, the two networks are trained together by minimizing the difference between the original time series and the reconstruction of said original time series, starting from random weights of both networks. The system obtains the required gradients by applying the chain rule to propagate the error derivatives back through the decoder network first and then through the encoder network. This system is referred to here as an autoencoder.

[0046] The unsupervised machine learning process described above for dynamic time warping (DTW) based time series detection can also be done supervised. According to the present invention, two execution variants of the learning strategy - supervised and unsupervised - can be applied to time series together with DTW. For example, two supervised learning methods - incremental learning and learning using priority rejection - can be distinguished in terms of execution variants. The incremental learning process is simple in conception, but usually requires a large set of time series to be matched. The learning process using priority rejection can effectively reduce the matching time, but usually slightly reduces the recognition accuracy. For the execution variant of unsupervised learning, in addition to the variants discussed above, an automatic learning method based on, for example, maximum matching learning and learning based on using priority and rejection can be used. The maximum matching learning disclosed here can be used to intelligently select appropriate time series for system learning. The effectiveness and efficiency of all three machine learning methods just proposed for DTW can be demonstrated using appropriate time series detection tests.

[0047] In the case of detecting first and / or second anomalies and significant items 101, 102, 101, 102, 10i associated with the digital twin respectively the twin items 101, 102, 101, 102, 10i, the measured dynamics or states as a function of time are transmitted as input data patterns to a machine learning unit and based on the output values of the machine learning unit the measurement parameters of the digital twin are adjusted by means of the electronic control device comprised by the digital platform 1, wherein the machine learning unit classifies the input patterns based on a learning pattern and generates corresponding metrology parameters. By additionally measuring structure / operation parameters, including measurement parameters for detecting physical properties of the twin items 101, 102, 101, 102, 10i by means of measurement devices, and / or item parameters by means of proprioceptive sensors or measurement devices, and / or environmental parameters by means of exteroceptive sensors or measurement devices, the machine learning unit can be adapted to the input patterns based on the measured time series, for example. For example, in addition to the measured time series of dynamics / states, one or more of the operation parameters and / or structure parameters and / or environmental parameters of the items 101, 102, 101, 102, 10i can be transmitted as input data patterns to the machine learning unit as a function of time. The machine learning unit can be implemented, for example, based on static or adaptive fuzzy logic systems and / or supervised or unsupervised neural networks and / or fuzzy neural networks and / or genetic algorithm-based systems. The machine learning unit can comprise, for example, a Naive Bayes classifier as a machine learning structure. The machine learning unit can be implemented, for example, based on supervised learning structures comprising, for example, a logistic regression and / or a decision tree and / or a support vector machine (SVM) and / or a linear regression as a machine learning structure. The machine learning unit can be implemented, for example, based on unsupervised learning structures comprising, for example, K-means clustering or K-nearest neighbors and / or dimensionality reduction and / or association rule learning. The machine learning unit can be implemented, for example, based on reinforcement learning results comprising, for example, Q-learning. The machine learning unit can be implemented, for example, based on ensemble learning comprising, for example, bagging (bootstrap aggregating) and / or boosting and / or random forests and / or stacking. Finally, the machine learning unit can be implemented based on neural network structures comprising, for example, a feedforward network and / or a Hopfield network and / or a convolutional neural network or a deep convolutional neural network.

[0048] The digital twin, i.e. the digital virtual copy of the twinned physical items 101, 102,..., 10i can be constantly updated and analyzed, for example by measuring data from their real counterparts, i.e. the twinned physical items 101, 102,..., 10i and / or from the physical environment surrounding them in their real physical items 101, 102,..., 10i. The digital platform 1 is able to react to the digital twin of the items 101, 102,... and 10i and the digital platform 1 can initiate automated analysis related to historical data, current data and predictions. Thus, the digital platform 1 can be able to predict what will happen in each case and the associated risks and thus be able to automatically propose actions and provide appropriate signaling. Assuming that the virtual twin itself or the digital platform 1 is linked by appropriate technical means, when technically so implemented, the virtual twin itself or the digital platform 1 can even take action on the technical means of their real world twinned items 101, 102,..., 10i respectively.

[0049] The recording and storing of measurement data and captured parameter values as playback data related to the project 101, 102,..., 10i enables the realization of a playback function according to an aspect of the present application by means of the data structure of the permanent storage device 10, i.e. the stream of measured parameters measured by sensors and / or measuring devices associated with the twin project 101, 102,..., 10i. A variant of the playback function with the digital platform 1 is intended as a specific embodiment of the system according to the present application. This specific embodiment can be realized with or without the use of the cases discussed above in which the project development optimization or prediction development function is used, i.e. with or without adjustment of the digital twin parameters by means of the electronic signaling system control device based on the output values of the machine learning unit or with or without adjustment or non-adjustment of the operating parameters / structural parameters / environmental parameters of the twin project 101, 102,..., 10i. In principle, the recording of the project parameters and the measurement values 10ii can be triggered by detecting a first event and / or a second time detected by its anomaly and importance within a time series. In such a playback embodiment, the analysis measurement data (as playback data or analysis measurement data) is provided to monitor the time period in the playback mode of the digital twin and / or the twin project 101, 102,..., 10i. By means of a component (BG) of the client comprising a time-shift retrieval for the analysis measurement data, for example by means of a time marker (time-based marker) or an event area displayed for selection by the digital platform 1 to the operating unit 2 or the deployment unit 3, a time-shift of the playback data to real time is selected, and for example by means of a request from the digital platform 1, a time-shift of the playback data to real time is requested. The digital platform 1 provides the requested time period to the operating or deployment unit 2 / 3, the digital platform 1 compiles the requested time portion of the playback data in the form of a multimedia data package and transmits said multimedia data package over the network to the component's client at the operating unit 2 or the deployment unit 3. The client unpacks the multimedia data package and displays it on the operating unit 2 or the deployment unit 3, for example on a monitor, etc. This component can be part of the digital platform 1, for example, implemented as part of the electronic system control device, or as a network component that can access the digital platform 1 or a system control device with an integrated digital platform 1 via the network. The time data sets within the time range that can be retrieved, for example including the events detected by the digital platform 1, can be highlighted on the GUI. The entire real-time time analysis data stream can be recorded, or only the time range of the analysis data stream, i.e. the playback data, in which the first event and / or the second event is detected by the digital platform 1. The embodiment according to the present application can also be realized in such a way that the operating unit or the deployment unit 2 / 3 can jump to any point in time in the past of the recorded analysis data stream, i.e. independently of the anomaly event detection.Furthermore, the operating unit or deployment unit 2 / 3 can retrieve the analysis measurement data from the stored replay data stream, for example, with a time delay over a certain time range, jump (forward) or jump (rewind) in the recorded data stream at a time x by a time range. In particular, further embodiments can be realized in such a way that the connected twin 101, 102,..., 10i or its digital twin can be set again by the electronic system control device into the precise operating mode or state with the same measurement parameters as in the detected event area. The digital twin and / or the connected twin 101, 102,..., 10i can thus be passed through the event area again in real time, for example, for testing, optimization or other verification purposes related to the project development.

[0050] Reference signs

[0051] 1 Digital networking platform

[0052] 10 Persistent storage device holding project data

[0053] 101, 102,..., 10i Project (P1,... P i ) and respective project data

[0054] 10i1 Relationship to operating unit S i , task T x and development unit E z assigned to project P y i

[0055] 10i2 Project submission of operating unit S x for project Pi

[0056] 10ii Project data

[0057] 11 Task selector

[0058] 111 Task database

[0059] 1111 Task T1

[0060] 1112 Task T2

[0061] ...

[0062] 111i Task T i

[0063] 112 Machine-based matching structure / matcher

[0064] 1121 Matching algorithm

[0065] 12 Operating unit access interface

[0066] 121 Submission database

[0067] 1211 Submission data for S1

[0068] 1212 Submission data for S2

[0069] ...

[0070] 121i Submission data for S i

[0071] 122 Submission analyzer

[0072] 13 Development unit access interface

[0073] 131 Authentication request

[0074] 132 Authentication verification

[0075] 14 Project manager module

[0076] 141 Access controller

[0077] 142 Project analyzer

[0078] 143 Task management module

[0079] 144 Collaboration module

[0080] 145 File management module

[0081] 146 Quoting module

[0082] 1461 Quoting service interface

[0083] 1462 Quoting server

[0084] 15 Authentication registrar

[0085] 151 Skill authentication data store

[0086] 1511 Skill authentication C1

[0087] 1512 Skill authentication C2

[0088] ...

[0089] 151i Skill authentication C i

[0090] 152 Authentication evaluation module

[0091] 16 Account module

[0092] 161 Communication database

[0093] 162 Profile database holding operational unit accounts 1621 Operational unit profiles

[0094] ​163 save deployment unit account profile database

[0095] 1631 deployment unit profile

[0096] 1632 authenticate skill approval

[0097] 164 billing / accounting module

[0098] 17 web server

[0099] 171 firewall

[0100] 172 router

[0101] 18 network interface

[0102] 19 dispatcher

[0103] 2 operating unit (production unit)

[0104] 21, 22,..., 2i operating unit (S1, S2,..., Si) i )

[0105] 2i1 skill approval of operating unit Si i

[0106] 2i2 network-enabled device of operating unit Si i

[0107] 2i21 network interface

[0108] 3 deployment unit

[0109] 31, 32,..., 3i deployment unit (E1, E2,..., Ei) i )

[0110] 3i1 skill approval of deployment unit Ei i

[0111] 3i2 network-enabled device of deployment unit Ei i

[0112] 3i21 network interface

[0113] 4 data transmission network

[0114] 41 global backbone network Internet

[0115] 5 secure cloud-based network

[0116] 51, 52,..., 5i private secure network

[0117] 5i1 controlled cloud-based network access implementation variant to secure network 5i ​​​​

[0118] 1 Digital networking platform

[0119] 1.1 Service task database

[0120] 1.11 First service task

[0121] 1.12 Second service task

[0122] 1.13 Third service task

[0123] 1.2 Service provider interface / technician interface

[0124] 1.21 First service provider / first technician

[0125] 1.22 Second service provider / second technician

[0126] 1.23 Third service provider / third technician

[0127] 1.3 Input database

[0128] 1.31 Input data from 1.21

[0129] 1.32 Input data from 1.22

[0130] 1.33 Input data from 1.23

[0131] 2.1 Evaluation unit

[0132] 2.11 First / skill level

[0133] 2.12 Second / further skill level

[0134] 2.13 Third / further skill level

[0135] 2.2 Training unit

[0136] 2.21, 2.22, 2.23 Training session

[0137] 3.1 Matching unit

[0138] 11.1 Intelligent algorithm

[0139] 3.2 Service provider profile database

[0140] 3.21 Service provider profile relating to assigned 1.11, 1.12, 1.13

[0141] 3.22 Service provider data (number, time required, cost...) relating to successfully completed 1.11, 1.12, 1.13

[0142] 3.23 Service provider data related to language skills (writing, reading)

[0143] 3.24 Service provider data related to home address

[0144] 3.25 Service provider data related to travel flexibility (car, train, plane, individual transport or public transport)

[0145] 3.26 Service provider data related to equipment

[0146] 3.27 Service provider data related to other training abilities 1.21, 1.22, 1.23 3 Buyer

[0147] 5 Communication module for bidirectional training communication

[0148] 5.1 Service task ordering interface

[0149] 6.1 Feedback input interface

[0150] 6.11 Feedback data

[0151] 6.12 Feedback connection

[0152] 7.1 Feedback input interface

[0153] 7.11 Feedback data

[0154] 8 Rating unit

Claims

1. A digital platform (1) for providing controlled, project-driven networked interaction between an operation unit (2) and a deployment unit (3), the operation unit (2) and the deployment unit (3) accessing the digital networking platform (1) via a data transmission network (4) through network enabling devices (2i2 / 3i2), each unit (2 / 3) having a unit account (1621 / 1631) assigned in a communication database (161) of the digital networking platform (1), the unit account (1621 / 1631) having associated authentication and authorization credentials (5i1) provided by the digital cross-network platform (1) for controlled network access (5i1) to a secure, cloud-based network (5 / 51, 52, ..., 5i), wherein, The items (101, 102, ..., 10i) stored in the persistent storage device (10) of the digital networking platform (1) include at least one allocation relationship (10i1) between operation units (2i), one or more deployment units (3i), and project tasks (101, 102, ..., 10i), and wherein the digital networking platform (1) includes: a network interface (18) of the digital platform (1) for providing network access to the secure, cloud-based network (5 / 51, 52, ..., 5i) via the data transmission network (4) for the operation units (21, 22, ..., 2i) and the deployment units (31, 32, ..., 3i) by means of the network enabling device (2i2 / 3i2), characterized in that, The items (101, 102, ..., 10i) stored in the persistent storage device (10) of the digital networking platform (1) are accessible only to the assigned units (2 / 3) via a dedicated secure network (51, 52, ..., 5i) to provide upload and download access to the item data (10ii) related to the items (101, 102, ..., 10i) and to share the item data (10ii) with other units (2 / 3) according to the allocation relationship (10i1); The project manager module (14) extracts project data (10ii) from the project submission (10i2) of the operation unit (2) requesting new projects (101, 102, ..., 10i), wherein the new projects (101, 102, ..., 10i) are generated by the project analyzer (142) in the project manager module (14) based on the project submission data (10i2), and wherein the associated project data (10ii) is stored in the persistent storage device (10); A task selector (11) selects one or more tasks (111i) to be executed by a deployment unit (3) from a task database (111) based on the project submission data (10i2), and a task manager module (143) selects one or more deployment units (3) based on the certification skill approval data (1632) of the certification skill (3i1) that can be executed by a specific deployment unit (3) according to the indication of the deployment unit (3), wherein the selected task (111i) is assigned to at least one selected deployment unit (3) to provide a relationship (10i1) associated with the new project (101, 102, ..., 10i); The task selector (11) includes a machine learning-based matching structure (112), wherein appropriate skill certifications (1511, 1512, ..., 151i) are determined by means of the machine learning-based matching structure (112) to perform a specific task (111i) associated with the project (10i), and wherein at least one deployment unit (31, 32, ...) is determined based on the determined required skill certifications (1511, 1512, ..., 151i) and additional triggering criteria. …, 3i), the additional triggering criteria include at least the location of the operation unit (21, 22, …, 2i) that submitted the data (1211, …, 121i) and the location of the available deployment unit (31, 32, …, 3i), wherein, in order to select the deployment unit (31, 32, …, 3i), the machine-based matching structure (112) determines the required skills and optional skills to perform the selected task (111i), wherein the required skills (1511, 1512) The optional skills (1511, 1512, ..., 151i) are necessary to complete the selected task (111i), and wherein the optional skills (1511, 1512, ..., 151i) include at least speaking language parameters and / or additional skill parameters indicating additional services or devices, and the deployment units (31, 32, ..., 3i) are filtered by the machine-based matching structure (112) for the required skills (1511, 1512, ..., 151i), while for the optional skills ( The deployment units (31, 32, ..., 3i) are assigned a scoring structure by the machine-based matching structure (112), wherein the scoring is based on maximum margin matching, taking into account the pricing parameter value and / or cost parameter value and / or travel time or geographical distance parameter value of the deployment units (31, 32, ..., 3i), and wherein the higher the score, the greater the likelihood that the deployment unit (31, 32, ..., 3i) will be selected and triggered by the machine-based matching structure and matcher (112). In order to generate the future state of projects (101, 102, ..., 10i) based on captured or measured project parameters, the matching structure (112) includes a digital copy of each project (101, 102, ..., 10i), and analyzes the digital copy of each project (101, 102, ..., 10i) based on a value time series generated for values ​​within a future time period for the future state to provide a measurement of the future state or operation of the twin real-world project (101, 102, ..., 10i), the measurement being related to the probability of occurrence of predefined events for possible development patterns of the project (101, 102, ..., 10i), wherein the detection of anomalies and significant events is triggered by a measurement deviation of project parameters for a single or set of operations and / or environments exceeding a defined threshold, and wherein the project is monitored and its development is predicted based on the time series, and the measurement of anomaly occurrence is quantified by triggering anomalous events within the time series. Access controller (141) generates a secure, cloud-based network (5 / 51, 52, ..., 5i) for the new project (101, 102, ..., 10i) to provide access to the new project (101, 102, ..., 10i) for the operation unit (2) and the selected deployment unit (3) represented by the relationship (10i1), wherein the relationship (10i1) is established between the paired units (2 / 3) before any communication between the paired units (2 / 3) is allowed.

2. The digital platform (1) according to claim 1, characterized in that, The rating is based on the best skill match.

3. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The scoring is also based on the platform's growth structure to give higher scores to deployment units (31, 32, ..., 3i) that use the platform (1) for tasks (111i) with little or no use.

4. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The machine-based matching structure (112) includes one or more machine-based learning structures (ML).

5. The digital platform (1) according to claim 4, characterized in that, The machine-based learning architecture (ML) includes hill climbing, genetic, or evolutionary machine-based learning architectures.

6. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The digital networking platform (1) includes a scheduler (19) for notifying one or more selected deployment units (31, 32, ..., 3i) and / or service providers associated with the appropriate deployment unit (31, 32, ..., 3i).

7. The digital platform (1) according to claim 6, characterized in that, The scheduler (19) includes an auction process to provide access to a specific task (111i) to the appropriate deployment unit (31, 32, ..., 3i).

8. The digital platform (1) according to any one of claims 1 to 2, characterized in that, Each allocation relationship (10i1) defines a hierarchical structure comprising at least two subgroups (2 / 3), wherein the first subgroup (2) comprises multiple operational units (21, 22, ..., 2i) defined by the unit account (1621) and the second subgroup (3) comprises multiple deployment units (31, 32, ..., 3i) defined by the unit account (1631).

9. The digital platform (1) according to claim 8, characterized in that, The secure cloud-based network access (5 / 51, 52, ..., 5i) provided via the data transmission network interface (18) for specific projects (101, 102, ..., 10i) is different for the operation units (21, 22, ..., 2i) of the first subgroup (2) and the deployment units (31, 32, ..., 3i) of the second subgroup (3).

10. The digital platform (1) according to claim 9, characterized in that, The operation unit (2) transmits the submitted data (10i2) to the digital networking platform (1) through the operation unit access interface (12), wherein the corresponding projects (101, 102, ..., 10i) are generated by means of the project analyzer (142) and the corresponding projects (101, 102, ..., 10i) are allocated to the persistent storage device (10).

11. The digital platform (1) according to claim 10, characterized in that, The task selector (11) selects one or more tasks (1111, 1112, ..., 111i) from the task database (111) based on the project data (10ii) associated with the project (101, 102, ..., 10i) for execution by the deployment unit (31, 32, ..., 3i).

12. The digital platform (1) according to claim 11, characterized in that, Based on the deployment unit profile (1631, 1632, ..., 163i), at least one deployment unit (31, 32, ..., 3i) is selected from the deployment unit profile database (163) to execute one or more selected tasks (1111, 1112, ..., 111i), wherein the deployment unit profile (1631, 1632, ..., 163i) is matched with one or more selected tasks (1111, 1112, ..., 111i) and related relationship data (10i1) is generated to define the relationship (10i1) between the new project (101, 102, ..., 10i) and the one or more tasks (1111, 1112, ..., 111i) to be executed.

13. The digital platform (1) according to claim 12, characterized in that, In order to select at least one deployment unit (31, 32, ..., 3i) from the deployment unit configuration file database (163) based on the deployment unit configuration file (1631, 1632, ..., 163i), each deployment unit configuration file (1631, 1632, ..., 163i) includes at least one certified task endorsement (1632) for one of the selectable tasks (1111, 1112, ..., 111i) in the task database (111).

14. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The network interface (18) is the web interface of the web server (17) of the digital platform (1), and the network enabling device (2i2 / 3i2) is a web enabling device, wherein the digital networking platform (1) includes optional productivity tools for interfacing with the projects (101, 102, ..., 10i), the productivity tools being accessible from the web interface and including at least a task manager module (143) and / or a collaboration module (144) and / or a file management module (145).

15. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The platform (1) further includes a quotation module (146) and / or a quotation server interface (1461) for periodically interacting with a quotation server (1462), retrieving financial information by means of the quotation module (146), and storing the retrieved financial information in the persistent storage device (10) allocated to the projects (101, 102, ..., 10i).

16. The digital platform (1) according to claim 15, characterized in that, The digital networking platform (1) includes a billing module (164) of an account module (16) for displaying and calculating financial information, which includes: financial account information and cost basis related to the unit (2 / 3) or unit account (1621 / 1631); and evaluation data related to the financial account information, which is generated based on the financial account information using retrieved task characteristic data associated with specific projects (101, 102, ..., 10i).

17. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The persistent storage device (10) is used in a distributed networked collaboration environment to store not only the project data (10ii), but also the relationships (10i1) and the project submission data (10i2) to provide a secure collaboration environment within the project company via a global backbone network (41).

18. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The authentication and authorization credentials include at least the username and password in the communication database (161).

19. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The digital networking platform (1) includes an authentication registrar (15) having a task authentication data repository (151) and an authentication evaluation module (152). The task authentication data repository (151) stores accessible task authentications (1511, 1512, ..., 151i) for each task (1111, 1112, ..., 111i) in the task database (111). The authentication evaluation module (152) includes an authentication evaluation process for approving the allocation of task authentications (1511, 1512, ..., 151i) for deployment units (31, 32, ..., 3i) for each task (1111, 1112, ..., 111i) in the task database (111). After approving a task authentication (1511, 1512, ..., 151i), a corresponding authentication task approval (1632) is assigned to the deployment unit (31, 32, ..., 3i) for the specific task.

20. The digital platform (1) according to any one of claims 1 to 2, characterized in that, By means of at least one sensor associated with a twin physical copy of the real-world project (101, 102, ..., 10i), structural parameters, operational parameters, and / or environmental state parameters (10ii) of the real-world project (101, 102, ..., 10i) are measured, monitored, and transmitted to the digital platform (1), wherein the at least one sensor comprises: one or more associated external sensing sensors or measuring devices for sensing external environmental parameters that physically affect the real-world project (101, 102, ..., 10i); and / or one or more proprioceptive sensors or measuring devices for sensing internal operational parameters or state parameters of the real-world project (101, 102, ..., 10i), wherein the sensor or measuring device includes a means for communication between the digital platform and the sensor or measuring device. One or more wireless or wired interfaces are configured, and data links can be established via the wireless or wired connections between the digital platform (1) and the sensors or measuring devices associated with the real-world assets to transmit the project status parameters (10ii) measured and / or captured by the sensors or measuring devices to the digital platform (1). The status parameters (10ii) are assigned to the digital twin representation, and the values ​​of the status parameters (10ii) associated with the digital twin representation are dynamically monitored and adaptively adjusted based on the transmitted parameters (10ii). The digital twin representation includes a data structure representing the state of each subsystem within a plurality of subsystems of the real-world project (101, 102, ..., 10i) to store the parameter values ​​(10ii) as a time series over a period of time. Using the digital platform (1), a data structure is generated based on a simulation application using cumulative damage modeling to represent the future state of each subsystem in a plurality of subsystems of the real-world project (101, 102, ..., 10i), as a value time series over a future time period. The cumulative damage modeling generates the effect of asset parameters of the twin real-world project (101, 102, ..., 10i) on the operation and / or environment over the future time period. Using the digital platform (1), the digital twin representation is analyzed to provide measurements of the future state or operation of the twin real-world project (101, 102, ..., 10i) based on value time series generated within the future time period. These measurements are related to the probability of occurrence of predefined physical events that physically affect the specific intensity or physical characteristics of the real-world project (101, 102, ..., 10i) or the probability of occurrence of predefined states of the real-world project (101, 102, ..., 10i). The frequency and severity of historical measurement data of the predefined physical events are used by a predictive machine learning module to establish physically measurable probability measurements of the occurrence or development of the predefined physical events using the measured historical time series data.

21. The digital platform (1) according to claim 20, characterized in that, For the avatar measurement of the evolving real-world measurement parameters (10ii), the control of the operation or state of the development of the real-world projects (101, 102, ..., 10i) is optimized or adjusted based on the measurement provided for the future state or operation of the twin real-world projects (101, 102, ..., 10i) and / or based on the value time series generated based on the values ​​within the future time period, so as to converge with the predefined operation and / or state asset parameters of the specific real-world projects (101, 102, ..., 10i), wherein, in the case of optimized operation control, the optimized operation control is generated to jointly and separately increase the specific operation performance standard of the operation unit (2) in the real-world projects (101, 102, ..., 10i) in the timely and future, or to reduce the measurement of the probability of occurrence associated with the operation or state of the real-world projects (101, 102, ..., 10i) within a specified probability range.

22. The digital platform (1) according to claim 20, characterized in that, The reduction in the measurement of the probability of occurrence of anomalies associated with the recorded time series of the real-world items (101, 102, ..., 10i) is based on the optimized or adjusted interaction between at least one operating unit (2) and the items (101, 102, ..., 10i) controlled by the digital platform (1).

23. The digital platform (1) according to any one of claims 1 to 2, characterized in that, The digital platform (1) also includes a cyber-physical twin structure that provides a digital representation of the evolution of real-world projects (101, 102, ..., 10i) for monitoring and / or predicting project development time series and quantifying the occurrence of anomalies by triggering anomalous events within the time series.

Images