Safety coach: risk mitigation of production processes

Abstract

A computer-implemented method for mitigating operation risks associated with one or more target chemical production and / or processing facilities, the method comprising: • obtaining a request for evaluating a configuration and / or operation of one or more target chemical production and / or processing facilities, wherein the request is associated with current configuration and / or operation data, wherein the current configuration and / or operation data is indicative a configuration and / or operation of the one or more target chemical production and / or processing facilities, • obtaining operation risk data comprising one or more operation risk(s) associated with the one or more target chemical production and / or processing facilities and one or more risk score(s) associated with the one or more operation risk(s), wherein the one or more risk score(s) are related to an occurrence of a deviation of the current configuration and / or operation data from target configuration and / or operation data • obtaining measure data comprising one or more measure(s) for mitigating the one or more operation risk(s) and one or more risk mitigation score associated with the one or more measure(s), wherein the one or more risk mitigation score(s) are related to a reduction of the one or more risk score(s) by applying the one or more measure(s) to the configuration and / or operation of the one or more target chemical production and / or processing facilities, • selecting at least one measure according to the one or more risk score(s) and the one or more risk mitigation score(s), • providing at least a part of the measure data related to the at least one selected measure for mitigating the at least one associated operation risk.

Description

[0001] 240150

[0002] SAFETY COACH: RISK MITIGATION OF PRODUCTION PROCESSES

[0003] TECHNICAL FIELD

[0004] The disclosure relates to a safe and error-free operation of chemical plants by operating instructions and to a method, in particular a computer-implemented method, for mitigating operation risks associated with one or more target chemical production and / or processing facilities, an apparatus for mitigating operation risks associated with one or more target chemical production and / or processing facilities, use of a measure obtained as described herein for mitigating an operation risk associated with a target chemical production and / or processing facility.

[0005] TECHNICAL BACKGROUND

[0006] Currently, configuration and / or operation of chemical plants have to fulfill a variety of requirements. This is facilitated by manually deriving and applying safety measures. Consequently, establishing a configuration and / or operation instructions takes a lot of time and resources and is prone to human-made errors. Hence, reliable and fast application of the target configuration and / or target operation instructions is desired.

[0007] SUMMARY

[0008] In an aspect, this disclosure relates to a method, in particular a computer-implemented method, for mitigating operation risks associated with one or more target chemical production and / or processing facilities, the method comprising: obtaining a request for evaluating a configuration and / or operation of one or more target chemical production and / or processing facilities, wherein the request is associated with current configuration and / or operation data, wherein the current configuration and / or operation data is indicative a configuration and / or operation of the one or more target chemical production and / or processing facilities, obtaining operation risk data comprising one or more operation risk(s) associated with the one or more target chemical production and / or processing facilities and one or more risk score(s) associated with the one or more operation risk(s), wherein the one or more risk score(s) are related to an occurrence of a deviation of the current configuration and / or operation data from target configuration and / or operation data, obtaining measure data comprising one or more measure(s) for mitigating the one or more operation risk(s) and one or more risk mitigation score associated with the one or more measure(s), wherein the one or more risk mitigation score(s) are related to a reduction of the one or more risk score(s) by applying the one or more measure(s) to the configuration and / or operation of the one or more target chemical production and / or processing 240150

[0009] 2 facilities, selecting at least one measure according to the one or more risk score(s) and the one or more risk mitigation score(s), providing at least a part of the measure data related to the at least one selected measure for mitigating the at least one associated operation risk.

[0010] In another aspect, it relates to an apparatus for mitigating operation risks associated with one or more target chemical production and / or processing facilities, the apparatus comprising: a processor configured for performing any one of the methods as described herein.

[0011] In another aspect, it relates to use of a measure obtained as described herein for mitigating an operation risk associated with a target chemical production and / or processing facility.

[0012] EMBODIMENTS

[0013] In the following, terminology as used herein and / or the technical field of the present disclosure will be outlined by ways of definitions and / or examples. Where examples are given, it is to be understood that the present disclosure is not limited to said examples.

[0014] These and other objects, which become apparent upon reading the following description, are solved by the subject matters of the independent claims. The dependent claims refer to embodiments of the invention.

[0015] Chemical plants have high safety requirements to ensure error-free production as errors in the operation provide hazardous risks to workers, equipment and environment. Followingly, configuration and / or operation of chemical production and / or processing facilities are checked thoroughly in highly complex and time-consuming processes. This task is typically performed manually by technical experts in the field because of long and complex data associated with requirements for an error-free configuration and / or operation of chemical production and / or processing facilities. Such data needs to be human-interpretable as this data is obtained from technical experts being the benchmark for safe configuration and / or operation of chemical production and / or processing facilities. Depending on the experience of the technical experts, different measures may be applied to eliminate operations risks achieving target safety requirements. Furthermore, a reliable reproduction of safety measure by workers of chemical production and / or processing facilities are required to ensure actual safe operation of chemical production and / or processing facilities. Here, a good balance between achieving lowest risks possible for incidents in chemical production and / or processing facilities and allowing for energy and material efficient production and / or processing of chemical products. 240150

[0016] 3

[0017] Risk scores allow for a quantification of a certainty that an event associated with a malfunction of the chemical production and / or processing facility may occur and a degree of severeness associated with the malfunction. Risk mitigation scores allow for a quantification of a reduction of risk scores when applying measures to the configuration and / or operation of the chemical production and / or processing facility. Thereby, operation risks and measures can be made comparable among each other. This enables to select targets with regard to safe operation of chemical production and / or processing facilities and achieve those targets without comprising on production and / or processing efficiency of the chemical production and / or processing facility. Selecting at least one measure according to the one or more risk score(s) and the one or more risk mitigation score(s) allows for a value-driven balancing of safety requirements and production and / or processing efficiency of the chemical production and / or processing facility. A consequence of good balancing the often highly complex safety procedures are simplified. Thereby, workers can more reliably apply safety measures. Furthermore, deciding in a score-driven manner allows to cluster risks and measures. In turn, larger sets of equipment can be judged to identify synergistic or antagonistic effects associated with the suggested measures. A holistic approach in mitigating operation risks can reduce safety measures needed as operation risks can be eliminated from the beginning. The effects of an early elimination can then propagate through the complex chemical production and / or processing facilities. Ultimately, selecting at least one measure according to the one or more risk score(s) and the one or more risk mitigation score(s) allows for scaling of risk mitigation throughout large chemical production and / or processing facilities and thus, a more reliable and safe operation of, in particular large, chemical production and / or processing facilities.

[0018] In an embodiment, the one or more operation risk(s) may be associated, in particular related to a configuration and / or operation of the one or more chemical production and / or processing facilities, in particular the one or more target chemical production and / or processing facilities. The one or more operation risk(s) may be related to at least a part of the equipment of the one or more target chemical production and / or processing facilities and / or at least a part of the operating instructions for controlling at least a part of the equipment of the one or more target chemical production and / or processing facilities. The equipment of the one or more target chemical production and / or processing facilities may include at least a part of machinery for processing and / or producing one or more chemical product(s). The one or more operation risk(s) may be related to a deviation of the configuration and / or operation of the one or more target chemical production and / or processing facilities from a target configuration and / or operation of the one or more target chemical production and / or processing facilities.

[0019] In an embodiment, the risk score(s) may quantify a number of events associated with a deviation of the current configuration and / or operation data from the target configuration and / or operation data. In particular the number of events may be a number of events within a predefined time interval. Risk score can be indicative of an impact associated with an operation risk and / or a certainty associated with an appearance of the operation risk. 240150

[0020] 4

[0021] In an embodiment, the risk mitigation score(s) may quantify a reduction of the risk score in relation to applying the one or more measure(s) to the one or more target chemical production and / or processing facilities.

[0022] In an embodiment, the configuration and / or operation of the one or more target chemical production and / or processing facilities may be related to an arrangement of equipment associated with the one or more target chemical production and / or processing facilities and / or an operation of the one or more target chemical production and / or processing facilities. The operation of the one or more target chemical production and / or processing facilities may be related to and / or may comprise one or more operating instructions for operating the one or more target chemical production and / or processing facilities.

[0023] In an embodiment, applying a measure to a configuration and / or operation of the one or more target chemical production and / or processing facilities may result in mitigating one or more operation risk(s) associated with the measure. In particular, applying the measure to the configuration and / or operation of the one or more target chemical production and / or processing facilities may result in approaching the target configuration and / or operation data associated with the one or more target chemical production and / or processing facilities. Applying the measure may result in changing the configuration and / or operation of the one or more target chemical production and / or processing facilities in particular towards the target configuration and / or operation data associated with the one or more target chemical production and / or processing facilities. In an embodiment, a measure may decrease a number of occurrences of the deviation of the current configuration and / or operation data from the target configuration and / or operation data within a predefined time interval.

[0024] In an embodiment, the one or more data-driven model may be pretrained data-driven model(s). The pretrained data-driven model(s) may be parametrized and / or trained based on unstructured data, in particular text data and optionally numerical data such as tabular data or image data. The pretrained data-driven model(s) may be configured to perform a plurality of task. The pretrained data-driven model(s) may be configured to perform the task according to the provided task instruction. Hence, the pretrained data-driven model may be configured to be provided with a plurality of different task instructions and / or provide a plurality of different types of output data upon receiving different task instructions. This allows to handle a plurality of use cases by one data-driven model. Such data-driven models are available and can directly be operated e.g. via API calls. The pretrained data-driven model(s) may be trained and / or parametrized based on unstructured data, in particular text data and optionally numerical data such as tabular data or image data. The unstructured data may be associated with a plurality of different tasks and / or task instructions. The pretrained data-driven model(s) may be configured to perform a plurality of task. The pretrained data-driven model(s) may be configured to perform the task according to the provided task instruction. Hence, the pretrained data-driven model(s) may be configured to be provided with a plurality of different task instructions and / or provide a plurality of different types of output data upon receiving different task instructions. 240150

[0025] 5

[0026] In an embodiment, the one or more data-driven model(s) may be finetuned data-driven model(s). The finetuned data-driven model(s) may be obtained by training pretrained data-driven model(s) configured to perform a plurality of tasks according to a plurality of task instructions. The finetuned data-driven model(s) may trained additionally on a training data set comprising a plurality of task instructions of one type and corresponding output data. The finetuned data-driven model may be trained additionally to provide output data of a predefined type according to the training data set. The finetuned data-driven model may be configured to be provided with a plurality of different task instructions and / or provide a plurality of different types of output data upon receiving different types of task instructions. Further, the finetuned data-driven model may be configured for providing one type of output data upon receiving one type of task instruction with a higher accuracy than providing other types of output data upon receiving other types of task instructions. The finetuned data-driven model(s) may be obtained by training a pretrained data-driven model based on historical verifying task instructions and corresponding indications on the verification of the identification of the operation risks. Additionally or alternatively, the finetuned data-driven model(s) may be obtained by training a pretrained data-driven model based on historical risk identification task instructions and corresponding one or more operation risks. Additionally or alternatively, the finetuned data-driven model(s) may be obtained by training a pretrained data-driven model based on historical measure determining task instructions and corresponding identified measures.

[0027] In an embodiment, the configuration and / or operation data may be related to, in particular may comprise at least parts of operating instructions for controlling at least a part of one or more target chemical production and / or processing facilities. The operating instructions may trigger and / or imitate producing and / or processing of one or more chemical product(s) by at least a part of the one or more target chemical production and / or processing facilities. The operating instructions may be provided to an operating engine of the one or more target chemical production and / or processing facilities. The operating instructions may be configured for triggering the operating engine of the one or more target chemical production and / or processing facilities. The current operating instructions may be operating instructions used for controlling at least a part of the one or more target chemical production and / or processing facilities, in particular at least once preceding the providing of the request for verifying the current operating data. The target operating instructions may be operating instructions used for controlling at least a part of the one or more chemical production and / or processing facilities, in particular at least once preceding the providing of the request for verifying the current operating data. The one or more chemical production and / or processing facilities may comprise the one or more target chemical production and / or processing facilities. The target operating instructions may operate the one or more chemical production and / or processing facilities associated with an absence of one or more operation risk(s). The target operating instructions may be associated with error-free operation of at least the part of the one or more chemical production and / or processing facilities. The target operating instructions may be verified. The target operating instructions may be historical operating instructions. 240150

[0028] 6

[0029] In an embodiment, the request may be associated with an indication of the one or more target chemical production and / or processing facilities and the current configuration and / or operation data may be retrieved based on the indication of the one or more target chemical production and / or processing facilities. By doing so, the user can provide request without detailing the configuration and / or operation of the one or more target chemical production and / or processing facilities. As the configuration and / or operation of chemical production and / or processing facility is typically highly complex and / or extensively, humans may be prone to errors in providing large datasets. Furthermore, the configuration and / or operation may be already available in a database. Hence, providing the configuration and / or operation data may be simplified by using already available data. Ultimately, this reduces computing resources used for inputting the configuration and / or operation data additionally and checking for errors in the configuration and / or operation data.

[0030] In an embodiment, obtaining the operation risk data may include providing a plurality of operation risks by a database and identifying the one or more operation risk(s) by retrieving the one or more operation risk(s) from the database. Retrieving the one or more operation risk(s) associated with the one or more target chemical production and / or processing facilities from the databased may include providing a request for retrieving the one or more operation risk(s) to the database. The database may be configured to provide the requested data upon receiving the request for retrieving the data. The request for retrieving the data may comprise the indication of the one or more target chemical production and / or processing facilities. The indication of the one or more target chemical production and / or processing facilities may be suitable for identifying the one or more target chemical production and / or processing facilities from the plurality of chemical production and / or processing facilities.

[0031] In an embodiment, the one or more operation risk(s) provided by the database and / or the plurality of operation risks may be provided to a data-driven model for verifying the identification together with verifying task instructions for verifying the identification of the one or more operation risk(s) associated with the one or more target chemical production and / or processing facilities and the configuration and / or operation data and optionally the plurality of operation risks other than the identified operation risk(s). The data-driven model may be configured to follow requests provided to the data-driven model. Providing the verifying task instructions to the data-driven model may trigger the data-driven model to generate an indication on a verification of the identification of the one or more operation risk(s).

[0032] By doing so, operation risks can be determined in a scalable manner. In turn, this enables timely reaction to potential operation risks and thus, can reduce impact of operation risks.

[0033] In an embodiment, obtaining the operation risk data may comprise obtaining an indication of the one or more operation risk(s) via a user interface, in particular provided by a technical expert associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities. 240150

[0034] 7

[0035] The indication of the one or more operation risk(s) may comprise a selection of the one or more operation risk(s) and / or the one or more operation risk(s). Preferably, the indication of the one or more operation risk(s) may be provided via a user interface in response to determining that the plurality of operation risks are associated with a plurality of chemical production and / or processing facilities other than the one or more target chemical production and / or processing facilities. In particular, the indication of the one or more operation risk(s) may be provided via a user interface in response to determining that the plurality of operation risks are associated with a plurality of chemical production and / or processing facilities of a different type and / or at a different location than the one or more target chemical production and / or processing facilities. By doing so, the technical expert can be consolidated upon identifying operation risks. As configuration and / or operation of chemical production and / or processing facilities is highly complex and requires highly skilled and experienced technical experts, reliability of the technical expert is typically higher and more tailored than a suggestion from a data-driven model. Hence, allowing the technical expert to be consolidated, in particular automatically, allows to improve the reliability of identifying operation risks. Therefore, operation risks can be more effectively resolved and thus, the configuration and / or operation of chemical production and / or processing facilities is improved.

[0036] In an embodiment, any one of the methods may further comprise receiving an indication of a verification of the one or more operation risk(s) via a user interface. Additionally or alternatively, the one or more operation risk(s) and the configuration and / or operation data may be displayed to verify the one or more operation risk(s) may be associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities. By doing so, the technical expert can be consolidated upon identifying operation risks. As configuration and / or operation of chemical production and / or processing facilities is highly complex and requires highly skilled and experienced technical experts, reliability of the technical expert is typically higher and more tailored than a suggestion from a computing apparatus. Providing a suggestion to the technical expert reduces the amount of data to be provided by the technical expert to a Boolean value indicative of whether the operation risk is identified correctly. Hence, this combines the advantage of improving the reliability of identifying operation risks while reducing the time needed for the technical expert to check. Therefore, the process can be scaled towards a higher number of request. Ultimately, this improves configuration and / or operation of chemical production and / or processing facilities in a large scale with high reliability.

[0037] In an embodiment, any one of the methods may further comprise receiving an indication of a verification of the one or more measure(s) via a user interface. Additionally or alternatively, the one or more measure(s) and the configuration and / or operation data may be displayed to verify the one or more measure(s) are associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities. By doing so, the technical expert can be consolidated upon identifying measures. As configuration and / or operation of chemical production and / or processing facilities is highly complex and requires highly skilled and experienced technical experts, reliability of the technical expert is typically higher and more tailored than a suggestion from a 240150

[0038] 8 computing apparatus. Providing a suggestion to the technical expert reduces the amount of data to be provided by the technical expert to a boolean value indicative of whether the operation risk is identified correctly. Hence, this combines the advantage of improving the reliability of identifying operation risks while reducing the time needed for the technical expert to check. Therefore, the process can be scaled towards a higher number of request. Ultimately, this improves configuration and / or operation of chemical production and / or processing facilities in a large scale with high reliability.

[0039] In an embodiment, obtaining the measure data may include identifying the one or more measure(s) associated with the one or more operation risk(s) by providing measure determining task instructions to a data-driven model. The data-driven model may be configured to follow task instructions. The measure determining task instructions may be related to, in particular include, the one or more identified operation risk(s), and the current configuration and / or operation data. The measure determining task instructions may further include a set of measures for mitigating a plurality of operation risks associated with the one or more target chemical production and / or processing facilities. The set of measures may be retrieved from a database comprising a plurality of measures associated with a plurality of chemical production and / or processing facilities. The measure determining task instructions may trigger the data-driven model to provide the one or more identified measure(s). By doing so, the measures may be obtained fast. Thus, this enables scaling of the methods described herein to large networks of chemical production and / or processing facilities, i.e. a Verbund site. Furthermore, the data-driven model can process structured and unstructured data. Hence, the data-driven model can process human-interpretable data. This allows the data-driven model to operate on the same data as the technical experts and thus, use available data sets to provide measures for mitigating operations risks.

[0040] In an embodiment, obtaining the operation risk data may include identifying the one or more operation risk(s) by providing risk identification task instructions to a data-driven model. The data-driven model may be configured to follow task instructions. The risk identification task instructions may be related to, in particular include, a plurality of operation risks associated with a plurality of chemical production and / or processing facilities and the configuration and / or operation data. The risk identification task instructions may trigger the data-driven model to provide the one or more operation risk(s) and / or a selection of the one or more operation risk(s). By doing so, the risk scores may be obtained fast. Thus, this enables scaling of the methods described herein to large networks of chemical production and / or processing facilities, i.e. a Verbund site.

[0041] In an embodiment, the measure data and / or operation risk data may comprise natural language and / or wherein the data-driven model may be trained to process natural language data. Processing natural language data by the data-driven model may refer to processing a numerical representation of the natural language data. The numerical representation of the natural language data may be obtained by mapping the natural language data to the numerical representation of the natural language data, e.g. by providing the natural language data to one or more embedding layer(s). The one or more embedding layer(s) may be configured to map data into a numerical representation of the data. By doing so, measure and / or operation risks may be available in a human-interpretable format. This enables technical experts in the configuration and / or operation of chemical production and / or processing facilities to validate and / or control mitigating operation risks by any one of the computer-implemented methods. Thereby, the resulting operation risks and measures are more reliable while obtained in a scalable manner.

[0042] In an embodiment, any one of the methods may further comprise providing a digital identifier suitable for retrieving one or more dataset(s) associated with the at least one measure data and / or operation risk data. Obtaining the measure data and / or the operation risk data may comprise retrieving the measure data and / or the operation risk data based on the digital identifier. The digital identifier may be a link to the one or more dataset(s), e.g. stored by a database. By doing so, the user, in particular the technical expert can access the datasets used for providing the measures and / or operation risks. Thereby, further information for implementing the measures or historical examples for implementation of the measures can be obtained. Especially where the measures and / or operation risks may be provided by a data-driven model, the provided measures may be subject to hallucination of the data- driven model. Hallucination is non-obvious to users receiving the measures. Hence, it is desired to allow identification of hallucination. By providing a link to the original data in the database, the validity of the measures and operation risks can be evaluated. This improves trust and enables human users to validate decisions from data-driven models. Ultimately, this increases reliability of mitigating operations risks by any one of the methods as described herein.

[0043] In an embodiment, obtaining the request for evaluating the configuration and / or operation of the one or more target chemical production and / or processing facilities may be triggered by determining that current configuration and / or operation data associated with the one or more target chemical production and / or processing facilities deviates from target configuration and / or operation data. By doing so, operations and / or configurations of chemical production and / or processing facilities can be evaluated upon determining deviations between target configuration and / or operation and current, i.e. actual, configuration and / or operation of chemical production and / or processing facilities. This allows to shorten time gaps between detecting operation risks and applying measures. By reducing this time gap, events in relation to the operation risks can be mitigated or even circumvented. This saves time, energy and material for reparation while providing higher safety for workers of the chemical production and / or processing facilities.

[0044] In an embodiment, the current configuration and / or operation data may comprise image data related to equipment of the one or more target chemical production and / or processing facilities. The image data may relate to at least one technical drawing associated with equipment of the one or more target chemical production and / or processing facilities. Typically, current configuration and / or operation data includes said technical drawings. These technical 240150

[0045] 10 drawings include all necessary information for identifying operation risks in a comprehensive data format. Processing such technical drawings to as current configuration and / or operation data prevents transformations of the comprehensive technical drawings into other formats potentially associated with higher amounts of data and / or less accurate than the technical drawings. Followi ngly , this improves reliability of mitigating operation risks and thus improves operation and / or configuration of chemical production and / or processing facilities.

[0046] In an embodiment, the one or more measure(s) may be further associated with one or more technical experts. Providing the one or more measure(s) may include providing an indication of the one or more technical experts. This allows to discuss details of implementing the provided measures, validate the recommended measures and / or tailor the measure to the one or more target chemical production and / or processing facilities by the technical expert.

[0047] In an embodiment, obtaining the measure data may comprise obtaining an indication of the one or more measure(s) via a user interface, in particular provided by a technical expert associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities. The indication of the one or more measure(s) may comprise the one or more measure(s), in particular the measure data, and / or a selection of the one or more measure(s). Preferably, the indication of the one or more measure(s) may be provided via a user interface in response to determining that the plurality of measures are associated with a plurality of chemical production and / or processing facilities other than the one or more target chemical production and / or processing facilities. In particular, the indication of the one or more measure(s) may be provided via a user interface in response to determining that the plurality of measures are associated with a plurality of chemical production and / or processing facilities of a different type and / or at a different location than the one or more target chemical production and / or processing facilities. By doing so, the technical expert can be consolidated upon identifying measures. As configuration and / or operation of chemical production and / or processing facilities is highly complex and requires highly skilled and experienced technical experts, reliability of the technical expert is typically higher and more tailored than a suggestion from a data-driven model. Hence, allowing the technical expert to be consolidated, in particular automatically, allows to improve the reliability of identifying operation risks. Therefore, measures can be more effectively resolved and thus, the configuration and / or operation of chemical production and / or processing facilities is improved.

[0048] In an embodiment, any one of the methods may further include providing the one or more risk score(s) and / or risk mitigation score(s) together with the at least one measure, in particular via a user interface. 240150

[0049] 11

[0050] In an embodiment, the one or more measure(s) and / or the measure data may be obtained from a plurality of measures associated with a plurality of chemical production and / or processing facilities and / or the one or more operation risk(s) may be obtained from a plurality of operation risks associated with the plurality of chemical production and / or processing facilities.

[0051] In an embodiment, the one or more operation risk(s) and / or the one or more measure(s) may be included into a database comprising a plurality of operation risks and / or measures in response to obtaining the one or more operation risk(s) and / or the one or more measure(s) via a user interface. By doing so, input from the technical experts can be integrated automatically into knowledge databases. This enables an improvement of the method for mitigating operation risks by allowing the system to learn from the technical experts. As a consequence, technical expert input may be limited to unseen and / or non-standard operation risks and already observed operation risks can be eliminated by using any one of the herein described methods. This saves time of the technical experts and accelerates providing the measures. This allows to shorten time gaps between detecting operation risks and applying measures. By reducing this time gap, events in relation to the operation risks can be mitigated or even circumvented. This saves time, energy and material for reparation while providing higher safety for workers of the chemical production and / or processing facilities.

[0052] In an embodiment, operation risk may refer to a difference and / or deviation between target operation and / or configuration and current operation and / or configuration. Target operation and / or configuration may be related to target operation instructions and / or a target arrangement of components associated with a chemical production and / or processing facility. The risk score may refer to a difference and / or deviation score. The risk mitigation score may refer to a difference and / or deviation mitigation score.

[0053] In an embodiment, mitigating may include eliminating and / or reducing occurrence of the one or more operation risk(s) to a predefined target occurrence.

[0054] In an embodiment, selecting the at least one measure according to the risk score and the risk mitigation score may include obtaining a target risk score associated with the chemical production and / or processing facility and selecting the at least one measure according to the risk score, the risk mitigation score and the target risk score. The target risk score may be associated with a target of an occurrence of the corresponding operation risk. The target risk score may be obtained from a database, via the user interface and / or from a look-up table according to the chemical production and / or processing facility associated with the request. The database and / or the look-up table may comprise two or more target risk scores associated with two or more chemical production and / or processing facilities. Target risk score may be associated with an occurrence of a deviation of the current configuration and / or operation data from target configuration and / or operation data. The target risk score may be 240150

[0055] 12 obtained according to a type of the operation risk. The type of the operation risk may be suitable for classifying operation risks. The types of the operation risk may be associated with a result of an occurrence of the respective operation risk. By doing so, the operation risks can be treated according to the consequences associated with the operation risks. This allows for a differentiated mitigation of operation risks. Thereby, resources for mitigating operation risks can be focused. Ultimately, this allows for a more reliable and safe operation of, in particular large, chemical production and / or processing facilities.

[0056] In an embodiment, the at least one measure may be selected according to a type of the measure. The type of the measure may be associated with an entity affected and / or targeted by the measure. For example, measures may comprise technical measures associated with technical equipment of the chemical production and / or processing facility. Additionally or alternatively, measure may comprise substitutional measure associated with substituting equipment or production and / or processing material associated with the chemical production and / or processing facility. Usually, substitution may be preferred over technical measures to mitigate, in particular eliminate operation risks. An example for this embodiment may be an application of the “STOP-Prinzip” common in the field of mitigating operation risks.

[0057] BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0058] In the following, the present disclosure is further described with reference to the enclosed figures. The same reference numbers in the drawings and this disclosure are intended to refer to the same or like elements, components, and / or parts.

[0059] FIG. 1 illustrates an embodiment of an operating system 106 of one or more target chemical production and / or processing facilities 102.

[0060] FIG. 2 illustrates an embodiment of a method for mitigating operation risk and / or controlling a chemical production and / or processing facility.

[0061] FIG. 3 illustrates an embodiment of a method for mitigating operation risk and / or controlling a chemical production and / or processing facility.

[0062] FIG. 4 illustrates examples of operations risks, types of chemical production and / or processing facilities, risk scores, measures and / or risk mitigation scores.

[0063] FIG. 5 illustrates an embodiment of a user interface for mitigating operation risks. 240150

[0064] 13

[0065] FIG. 6 illustrates an embodiment of a transformer architecture.

[0066] DETAILED DESCRIPTION

[0067] The following embodiments are mere examples for implementing the method, the system or application device disclosed herein and shall not be considered limiting.

[0068] The following embodiments are mere examples for implementing the method, the system or application device disclosed herein and shall not be considered limiting.

[0069] FIG. 1 illustrates an embodiment of an operating system 106 of one or more target chemical production and / or processing facilities 102.

[0070] One or more target chemical production and / or processing facilities 102 may be configured for producing one or more chemical products 104 from raw material 124. The one or more target chemical production and / or processing facilities 102 may comprise a plurality of components. For example, the components may include heat exchangers, cooling unit, filter, pumps, valves, separation chambers or the like. The one or more components of the one or more target chemical production and / or processing facilities 102 may be in connection with an interface 108. Preferably, the one or more target chemical production and / or processing facilities 102 may comprise one or more mean(s) for displaying at least parts of the data provided by the interface 108. The interface 108 may be configured to provide, in particular receive a request for mitigating operation risks associated with the one or more target chemical production and / or processing facilities 102. The interface 108 may provide the request to a model engine 110. The model engine 1 10 may be configured to identify operation risks associated with the one or more target chemical production and / or processing facilities 102, in particular by utilizing data available via one or more data source(s) 114. The one or more data source(s) 114 may comprise current configuration and / or operation data associated with the one or more target chemical production and / or processing facilities 102. The current configuration and / or operation data may allow to evaluate the configuration and / or operation of the one or more target chemical production and / or processing facilities 102 as suggested and / or operated. Further, the one or more data source(s) 114 may comprise a plurality of risk scores associated with a plurality of different chemical production and / or processing facilities. The model engine 110 may be configured to identify one or more operation risk(s) associated with the one or more target chemical production and / or processing facilities 102 as described in further detail within the context of FIG. 2. Based on the identified operation risks, associated risk scores may be determined. For example, the one or more data source(s) 1 14 may comprise a plurality of risk scores and the model engine 110 may identify, in particular retrieve the risk score(s) associated with the one or more identified operation risk(s). Additionally or alternatively, the one or more data source(s) 114 may comprise a relation 240150

[0071] 14 between the operation risks and the risk scores and the model engine 110 may determine the risk scores associated with the one or more identified operation risk(s) based on the relation. Based on the identified operation risks and risk scores, the model engine 110 may be further configured to identify one or more measure(s) associated with the one or more identified operation risk(s), in particular to eliminate the one or more operation risk(s). Therefore, the model engine 1 10 may be configured to determine risk mitigation scores. For example, the one or more data source(s) 114 may comprise a plurality of risk mitigation scores and the model engine 110 may identify, in particular retrieve the risk mitigation score(s) associated with the one or more identified measure(s). Additionally or alternatively, the one or more data source(s) 114 may comprise a relation between the measures and the risk mitigation scores and the model engine 110 may determine the risk mitigation scores associated with the one or more identified measure(s) based on the relation.

[0072] The identified operation risks, measures and corresponding risk scores and risk mitigation scores may be provided via the interface 108, e.g. a user interface and / or may be displayed to workers associated with the one or more target chemical production and / or processing facilities 102 as described in further detail in the context of

[0073] FIG. 2.

[0074] The operating system 106 may be configured to perform any one of the methods as described in the context of FIG. 2 and / or FIG. 3.

[0075] FIG. 2 illustrates an embodiment of a method for mitigating operation risk and / or controlling a chemical production and / or processing facility, in particular operation and / or configuration and / or operation of a chemical production and / or processing facility.

[0076] A request for evaluating an operation of one or more target chemical production and / or processing facilities may be provided 202. The request may be associated with an indication on the one or more target chemical production and / or processing facilities and / or an indication on the one or more target chemical production and / or processing facilities. The request may comprise unstructured data. In particular, the request may be human-interpretable, e.g. the request may comprise natural language. The request may be provided by a user and / or via a user interface.

[0077] Providing the request may be triggered by detecting an incident or a malfunction associated with in the one or more target chemical production and / or processing facility and / or a deviation of the current operation of the one or more target chemical production and / or processing facilities from a target operation. Detecting the incident may include determining one or more parameters associated with the one or more chemical production and / or processing facilities being within an operating risk range, preferably for a time interval being equal or larger than an operation risk time interval. 15

[0078] The request may be suitable for identifying the one or more target chemical production and / or processing facilities. For example, the request may comprise an identifier associated with the one or more target chemical production and / or processing facility. The identifier may be a digital identifier and / or may point to data associated with the one or more target chemical production and / or processing facilities.

[0079] In an embodiment, the request may comprise current configuration data and / or operation data. The configuration and / or operation data may be relate to a configuration and / or operation of the one or more target chemical production and / or processing facilities. The configuration and / or operation of the one or more target chemical production and / or processing facilities may be related to an arrangement of equipment associated with the one or more target chemical production and / or processing facilities and / or an operation of the one or more target chemical production and / or processing facilities. The operation of the one or more target chemical production and / or processing facilities may be related to and / or may comprise one or more operating instructions for operating the one or more target chemical production and / or processing facilities.

[0080] In an embodiment, the current configuration and / or operation data may comprise image data and / or text data. In particular, the image data may comprise a technical drawing associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities.

[0081] In an embodiment, current configuration and / or operation data associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities may be provided 204, in particular received, based on the indication on the one or more target chemical production and / or processing facilities. The current configuration and / or operation data may be provided and / or stored by a database. The database may be configured to provide the current configuration and / or operation data, in particular upon receiving a request for providing the current configuration and / or operation data. In particular, the request for providing the current configuration and / or operation data may comprise the identifier associated with the one or more target chemical production and / or processing facilities. The database may further comprise current configuration and / or operation data associated with a plurality of chemical production and / or processing facilities, in particular including the one or more target chemical production and / or processing facilities. The one or more data source(s) 114 may comprise the database associated with the current configuration and / or operation data.

[0082] A plurality of operation risks associated with operating the one or more chemical production and / or processing facilities may be provided 206. The plurality of operation risks may be provided by a database, in particular by the database associated with the current configuration and / or operation data. The database may relate the plurality of operation risks with the chemical production and / or processing facilities. Hence, based on the indication of one or more target chemical production and / or processing facilities, i.e. the identifier associated with the one or more 16 target chemical production and / or processing facilities, the plurality of operation risks may be provided. Examples of operation risks may be seen in FIG. 4.

[0083] One or more operation risk(s) associated with the one or more target chemical production and / or processing facilities may be identified based on the plurality of operation risks and the current configuration and / or operation data 208. The one or more operation risk(s) may be identified by providing identifying task instructions to a data- driven model for identifying the one or more operation risk(s). The identifying task instructions may comprise the plurality of operation risks and the current configuration and / or operation data. The data-driven model may be configured to follow the task instructions. The data-driven model may be a data-driven model as described in the context of FIG. 6. In particular, where the current configuration and / or operation data may comprise image data, the data-driven model may be associated with an image embedding configured to map the configuration and / or operation data to a structured and / or numerical representation of the configuration and / or operation data. Further, identifying the one or more operation risk(s) may comprise receiving the one or more operation risk(s) via a user interface, in particular operated by a human expert associated with operating the one or more target chemical production and / or processing facilities. The user interface may provide the current configuration and / or operation data and the plurality of operation risks to the human expert and / or the current configuration and / or operation data and the plurality of operation risks may be displayed to the human expert. This may be in particular advantageous where the plurality of operation risks may be independent of the current configuration and / or operation data associated with the one or more target chemical production and / or processing facilities.

[0084] In an embodiment, identifying the one or more operation risk(s) may comprise selecting the one or more operation risk(s) from the plurality of operation risks according to a similarity search. The similarity search may include matching structured representations of the plurality of operation risks and a structured representation of the current configuration and / or operation data. Additionally or alternatively, identifying the one or more operation risk(s) may comprise providing a request for retrieving the one or more operation risk(s) from a database comprising the plurality of operation risks.

[0085] If the one or more identified operation risk(s) may be different from the plurality of operation risks, the one or more identified operation risk(s) may be included into the plurality of operation risks, e.g. by entering the one or more operation risk(s) into the database comprising the plurality of operation risks.

[0086] A relation between the plurality of operation risks and risk scores associated with the operation risks may be provided 210. The relation may be suitable for obtaining the risk scores from the operation risks. In an embodiment, the relation may comprise one or more mathematical formula. The mathematical formula may be configured to relate at least parts of the operation risks to risk scores. Additionally or alternatively, the relation may be a data representation indicative of an associated between the operations risks and the risk scores, i.e. a 240150

[0087] 17 look-up table and / or a database comprising a relation and / or the association between the operation risks and the risk scores. The risk score may be indicative of an impact associated with an operation risk and a certainty associated with an occurrence of the operation risk.

[0088] The one or more risk score(s) associated with the one or more operation risk(s) may be determined 212. Determining the one or more risk score(s) may comprise applying the one or more mathematical formula to the one or more operation risk(s) and / or identifying the one or more risk score(s) associated with the one or more operation risk(s), e.g. via the database. In an embodiment, a plurality of risk scores may be associated with the plurality of operation risks. By identifying the one or more operation risk(s), the associated risk scores may be determined. Examples of risk scores may be seen in FIG. 4.

[0089] In an embodiment, the one or more operation risk(s) and / or the one or more risk score(s) may be provided e.g. via a user interface. Additionally or alternatively, an identifier associated with the one or more operation risk(s) and / or the one or more risk score(s) may be provided. This allows technical experts to validate the retrieval by the computing system.

[0090] A plurality of measures associated with at least a part of the plurality of operation risks may be provided 214, e.g. by a database and / or a table. The database may be the database comprising the operation risks. Hence, the database may comprise a plurality of operation risks and associated with at least a part of the operation risks a plurality of measures. The measures may be suitable for and / or configured to eliminate the one or more identified operation risk(s). Applying the measures to the one or more target chemical production and / or processing facilities may result in mitigating the one or more operation risk(s). Examples for measures can be seen in FIG. 4.

[0091] One or more measure(s) associated with the one or more operation risk(s) may be identified based on the plurality of measures associated with the one or more operation risk(s) and the one or more operations risk(s) 216. The one or more measure(s) may be identified analogous to identifying the one or more operation risk(s). Hence, identifying the one or more measure(s) may include using the data-driven model, providing the identified one or more measure(s) by the database including the plurality of measures and / or applying a similarity search. Further, identifying the one or more measure(s) may comprise receiving the one or more measure(s) via a user interface, in particular operated by a technical expert associated with operating the one or more target chemical production and / or processing facilities. The plurality of measures and the one or more identified operation risks may be displayed to the technical expert. Additionally or alternatively, identifying the one or more measure(s) may include providing the request and / or the one or more identified operation risk(s) to a technical expert associated with the operation of the one or more target chemical production and / or processing facilities. Additionally or alternatively, identifying the one or more measure(s) may include providing an indication of the technical expert associated with the one or more target chemical production and / or processing facilities. By doing so, the technical expert can directly be consultated as the technical expert is the benchmark for evaluating safe operation of chemical production and / or processing facilities. Hence, this improves reliable operation of the one or more target chemical production and / or processing facilities.

[0092] If the one or more identified measure(s) may be different from the plurality of measure(s), the one or more identified measure(s) may be included into the plurality of measures, e.g. by entering the one or more measure(s) into the database comprising the plurality of measures.

[0093] In an embodiment, the one or more identified measure(s) may be provided via a user interface and / or may be displayed to the technical expert associated with the operation of the one or more target chemical production and / or processing facilities. A verification of the one or more identified measure(s) may be provided via the user interface, in particular by the technical expert. Upon verifying the identified measure(s), the one or more measure(s) may be provided, e.g. via a user interface and / or displayed to a worker associated with the one or more target chemical production and / or processing facilities.

[0094] A relation between the plurality of measures, in particular the one or more measure(s), and risk mitigation scores associated with the measures, in particular one or more risk mitigation score(s) associated with the one or more measure(s) may be provided 218. The relation between the plurality of measures and the risk mitigation score may be analogous to the relation between the operation risks and the risk scores.

[0095] One or more risk mitigation score(s) associated with the one or more identified measure(s) may be determined 220. The one or more risk mitigation score(s) may be determined analogous to the one or more risk score(s).

[0096] At least one measure of the one or more identified one or more measure(s) may be selected according to the one or more risk mitigation score(s) and the one or more risk score(s) 222. To allow for a well-functioning of the one or more target chemical production and / or processing facilities, the risk mitigation score may be required to compensate for the risk scores, in particular to arrive at a target safety score. Where the risk score may be of at a certain level, the risk mitigation score may be at a similar level. Additionally or alternatively, a target safety score may be provided. The risk mitigation score may be required to reduce the risk score to the target safety score, i.e. the current safety score may be within a range characterized by the target safety score. In particular, the current safety score may be equal or below the target safety score. The current safety score may be a combination of the risk mitigation score and the risk score, e.g. a multiplication and / or subtraction. Examples of the risk scores and risk mitigation scores can be seen in FIG. 4. The at least one selected measure may be associated with the risk mitigation score that compensates the associated risk score, i.e. the at least one selected measure may be associated with a target risk mitigation score. The target risk mitigation score may be obtained based on and / or may be characterized by the risk score. A high risk score may result in a high target risk mitigation score. If there 240150

[0097] 19 are two or more measures with a target risk mitigation score, i.e. with a risk mitigation score within a predefined mitigation range, the at least one selected measure may comprise the two or more measures or the measure associated with a subrange of the predefined mitigation score may be selected. By doing so, suitable measures for operating the one or more target chemical production and / or processing facilities may be selected. Said measures allow to compensate for the operations risks. Hence, applying the measures to the one or more target chemical production and / or processing facilities may result in an error-free operation with respect to the one or more identified operation risk(s).

[0098] In an embodiment, selecting the at least one measure according to the risk score and the risk mitigation score may include obtaining a target risk score associated with the chemical production and / or processing facility and selecting the at least one measure according to the risk score, the risk mitigation score and the target risk score. The target risk score may be associated with a target of an occurrence of the corresponding operation risk. The target risk score may be obtained from a database, via the user interface and / or from a look-up table. The database and / or the look-up table may comprise two or more target risk scores associated with two or more chemical production and / or processing facilities.

[0099] In an embodiment, selecting the at least one measure may include receiving a selection of the at least one measure via a user interface, in particular provided by the technical expert associated with the operation of the one or more target chemical production and / or processing facilities, upon providing the one or more identified measure(s) and associated risk mitigation score(s) and further the one or more risk score(s) associated with the one or more operation risk(s) related to the one or more identified measure(s).

[0100] In another embodiment, the at least one selected measure, preferably the one or more identified measure(s) and an indication of the selection of the at least one selected measure and optionally the associated risk mitigation scores, may be provided via a user interface, in particular to be verified by the technical expert associated with the one or more target chemical production and / or processing facilities. Upon providing the at least one selected measure, an indication of a verification of the at least one selected measure may be provided by the technical expert and / or via the user interface. This allows to check the determined measure by the technical expert before investing resources such as time and material to stop the operation of the one or more target chemical production and / or processing facilities and apply the at least one selected measure. This improves the reliability of the selected measures to allow for an error-free and safe operation of the one or more target chemical production and / or processing facilities.

[0101] The at least one selected measure may be provided 224, e.g. via a user interface and / or may be displayed to a worker associated with the one or more target chemical production and / or processing facilities. The at least one 240150

[0102] 20 selected measure may be provided in response to receiving the selection of the at least one measure and / or an indication of a verification

[0103] FIG. 3 illustrates an embodiment of a method for mitigating operation risk and / or controlling a chemical production and / or processing facility.

[0104] FIG. 3 details the options for identification and / or verification of operation risks and / or measures in the method as described in the context of FIG. 2.

[0105] FIG. 4 illustrates examples of operations risks, types of chemical production and / or processing facilities, risk scores, measures and / or risk mitigation scores.

[0106] FIG. 5 illustrates an embodiment of a user interface for mitigating operation risks and / or operating one or more chemical production and / or processing facilities.

[0107] FIG. 6 illustrates an embodiment of a data-driven model, i.e. a transformer encoder comprising an encoder input 688, one or more encoder block(s) 686 and an encoder output 676 and / or a transformer decoder comprising a decoder input 694, one or more decoder block(s) 690 and a decoder output 692 and / or a transformer encoderdecoder.

[0108] The transformer encoder comprises an encoder input 678, one or more encoder blocks 674, 614 and an encoder output. A plurality of transformer encoder architectures are available in the art such as the bi-directional encoder representations from transformers (BERT). The input data may be received at the encoder input 678. The input data may comprise at least one of text data, numerical data, tabular data, image data or the like. Where the input data may comprise one of text data, numerical data, tabular data, image data or the like, input embedding of a type corresponding to the type of input data may be applied. Hence, the input embedding may be configured to map text data, numerical data, tabular data, image data or a combination thereof to a numerical representation of the input data. An example of input embedding associated with text data may be a continuous bag-of-words-model (CBOW). Additionally or alternatively, Word2Vec may be used for representing input data. Upon receiving the input data, the input data may be tokenized via a vocabulary associated with a preselection of elements of expected input data. Applying the input embedding may comprise mapping the input data, in particular the two or more elements of the input data to a numerical represenation of the input data, preferably of a predefined size e.g. via padding. Further, the encoder input 678 may apply positional encoding 604. For example, the positional factor pposmay be obtained based on the following equation:

[0109] 10000 7 240150

[0110] 10000 7 where pos may refer to the position of the element within the sequence, i may refer to the dimension associated with the input embedding and d may refer to the dimension of the data-driven model. Alternatively, the positional encoding may be based on rotary positional embeddings (RoPE). The embedded input data may be processed by the encoder block. The embedded input data may be provided to the layer normalization 608 by a residual connection. Multi-head self-attention 606 may be applied to the embedded input data. The embedded input data may serve as query Q, key K and value V with respect to the self-attention operation. For improving the efficiency, multiple heads are used to apply the filter according to the following equation: head i = Attention(QWiQ, KW* , 7VK ) with parameter matrices WLQe IR.dxd<2, W(Ke Rdxdfc, W,ve ]R.dxdv where I may refer to the number of heads, dv, dKand dQmay refer to the dimensions of the value, key and query.

[0111] The result of the two or more head may be concatenated according to the following equation: MultiHead(Q, K, V) = Concat head l, .. , headhW° where IVOe ^hdvxdanc|may number of heads. This may result in a context tensor. After the multi-head self-attention 606 layer normalization 608 may be applied based on the context tensor and / or the embedded input data from the residual connection. The so-obtained tensor may be passed to a feed-forward layer 610 again followed by layer normalization 612 based on the residual connection to the context tensor and / or the output of the feed-forward layer 610.

[0112] The encoder output 676 may comprise a linear layer 634 and a softmax layer 636. The encoder output may be configured to map the concatenated numerical representation of the two or more elements and the relation between the two or more elements to a numerical representation of the output data. Additionally or alternatively, a decoding model may be used to map the concatenated numerical representation of the two or more elements and the relation between the two or more elements to a numerical representation of the output data. The decoding model may be trained to relate a numerical represenation of data of a type according to the input data.

[0113] The transformer decoder comprises a decoder input 684, one or more decoder blocks 680, 628 and a decoder output 692. A plurality of transformer decoder architectures are available in the art such as the generative pretrained transformers (GPT). In contrast to the transformer encoder, the transformer decoder may perform masked multi-head self-attention 620 by additionally masking a part of the embedded input data associated with elements later in the sequence than the element to be generated. Additionally or alternatively, the part of the input data associated with elements later in the sequence than the element to be generated may not be received and / or transformed into the embedded input data.

[0114] The transformer encoder-decoder may comprise a combination of the transformer encoder and transformer decoder wherein the context tensor obtained from the encoder block may be used for the multi-head self-attention 664 operation in at least one decoder block 690. 240150

[0115] 22

[0116] Input data to the data-driven model may comprise image data. The encoder input and / or decoder input may comprise one or more linear projection layer(s) for a linear projection of a sequence of two or more partial images. This may result in changing the dimension of the one or more received images. Furthermore, positional embedding may be applied to the sequence, preferably by passing the sequence of one or more images and / or partial images through the one or more linear projection layer(s). Additionally or alternatively, the input data may comprise tabular data. Input embeddings for tabular input data may comprise a token embedding, a positional embedding, a column embedding, a row embedding or a combination thereof.

[0117] In an example, the data-driven model may comprise a Mamba block. A Mamba architecture may enhance inference speed in relation to a transformer based model. Mamba block may be based on a selective space state sequence model. A selective state space layer may be a linear recurrent network that selectively process data based on the input token, which may allow to focus on relevant data and discard irrelevant data. For instance in each step a separate weight vector may be determined based on the respective input token. The determined weight vector may then be used in a selective scan. A selective state space layer may be used in a convolutional mode e.g. for parallelizable training and a recurrent mode for near-constant time generation of output data. A state space operation may be based on solving the state and output equations, wherein a state equation may describe how a state changes based on how the input influences the state and an output equation may describe how the state is translated to the output. Further how the input influences the output may be represented by a learnable linear transformation used in a learnable skip connection. An example of the architecture of a mamba block may be found in “Mamba: Linear-Time Sequence Modeling with Selective State Spaces” by Albert Gu and Tri Dao arXiv:2312.00752v2 [cs.LG] 31 May 2024, , which is incorporated herein by reference.

[0118] In an example, any one of the data-driven models may further comrpise a mixture of experts block. The mixture of experts block may be decoder blocks wherein the feed-forward layer may be exchanged for a gating network and a number of parallel feed-forward layers, wherein the gating network may switch between the feed-forward layers depending on the input. This may allow leveraging advantages of the different architectures.

[0119] The data-driven model may generate one token per timestep upon processing the input data and optionally all output tokens already produced. The training data set may comprise a plurality of sequences comprising a plurality of elements. During the training of the data-driven model, sequences of the training data set may be provided to the data-driven model and one or more elements may be generated based on the sequences of the training data set one by another. The elements generated based on the sequences may follow the elements of the parts of sequences the data-driven model may have been provided with. The generated one or more elements may be compared to the one or more elements following the at least a part of the sequences provided to the data- driven model in order to adapt the parameters of the data-driven model depending on a deviation of the predefined token and the generated token.

[0120] The present disclosure has been described in conjunction with preferred embodiments and examples as well.

[0121] However, other variations can be understood and effected by those persons skilled in the art and practicing the 240150

[0122] 23 claimed subject-matter, from the studies of the drawings, this disclosure and the claims. Notably, in particular, the any steps presented can be performed in any order, i.e. the present disclosure is not limited to a specific order of these steps. Moreover, it is also not required that the different steps are performed at a certain place or at one node of a distributed system, i.e. each of the steps may be performed at different nodes using different equipment / data processing.

[0123] As used herein ..determining" also includes ..initiating or causing to determine", “generating" also includes ..initiating and / or causing to generate" and “providing” also includes “initiating or causing to determine, generate, select, send and / or receive”. “Initiating or causing to perform an action” includes any processing signal that triggers a computing node or device to perform the respective action.

[0124] In the claims as well as in the description the word “comprising” or “including” or similar wording does not exclude other elements or steps and shall not be construed limiting to the elements or steps lined out. The indefinite article “a” or “an” does not exclude a plurality. A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that certain measures are recited in the mutual different dependent claims does not indicate that a combination of these measures cannot be used in an advantageous implementation or further elements may be included.

[0125] Providing in the scope of this disclosure may include any interface configured to provide data. This may include an application programming interface, a human-machine interface such as a display and / or a software module interface. Providing may include communication of data or submission of data to the interface, in particular display to a user or use of the data by the receiving entity. Providing may include receiving. Obtaining may include providing, determining and in particular receiving.

[0126] Various units, circuits, entities, nodes or other computing components may be described as “configured to” perform a task or tasks. Configured to shall recite structure meaning “having circuitry that” performs the task or tasks on operation. The units, circuits, entities, nodes or other computing components can be configured to perform the task even when the unit / circuit / component is not operating. The units, circuits, entities, nodes or other computing components that form the structure corresponding to “configured to” may include hardware circuits and / or memory storing program instructions executable to implement the operation. The units, circuits, entities, nodes or other computing components may be described as performing a task or tasks, for convenience in the description. Such descriptions shall be interpreted as including the phrase “configured to.” Any recitation of “configured to” is expressly intended not to invoke 35 U.S.C. § 112(f) interpretation.

[0127] In general, the methods, apparatuses, systems, computer elements, nodes or other computing components described herein may include memory, software components and hardware components. The memory can include 24 volatile memory such as static or dynamic random-access memory and / or nonvolatile memory such as optical or magnetic disk storage, flash memory, programmable read-only memories, etc. The hardware components may include any combination of combinatorial logic circuitry, clocked storage devices such as flops, registers, latches, etc., finite state machines, memory such as static random-access memory or embedded dynamic random-access memory, custom designed circuitry, programmable logic arrays, etc.

[0128] Any disclosure and embodiments described herein relate to the methods, the systems, devices, the computer program element lined out above and vice versa. Advantageously, the benefits provided by any of the embodiments and examples equally apply to all other embodiments and examples and vice versa.

[0129] All terms and definitions used herein are understood broadly and have their general meaning.

Claims

24015025CLAIMSWhat is claimed is:1 . A method, in particular a computer-implemented method, for mitigating operation risks associated with one or more target chemical production and / or processing facilities, the method comprising: obtaining a request for evaluating a configuration and / or operation of one or more target chemical production and / or processing facilities, wherein the request is associated with current configuration and / or operation data, wherein the current configuration and / or operation data is indicative a configuration and / or operation of the one or more target chemical production and / or processing facilities, obtaining operation risk data comprising one or more operation risk(s) associated with the one or more target chemical production and / or processing facilities and one or more risk score(s) associated with the one or more operation risk(s), wherein the one or more risk score(s) are related to an occurrence of a deviation of the current configuration and / or operation data from target configuration and / or operation data, obtaining measure data comprising one or more measure(s) for mitigating the one or more operation risk(s) and one or more risk mitigation score associated with the one or more measure(s), wherein the one or more risk mitigation score(s) are related to a reduction of the one or more risk score(s) by applying the one or more measure(s) to the configuration and / or operation of the one or more target chemical production and / or processing facilities, selecting at least one measure according to the one or more risk score(s) and the one or more risk mitigation score(s), providing at least a part of the measure data related to the at least one selected measure, in particular for mitigating the at least one associated operation risk.

2. The method of claim 1 , wherein obtaining the operation risk data comprises obtaining an indication of the one or more operation risk(s) via a user interface, in particular provided by a technical expert associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities.

3. The method of claim 1 or 2, further comprising receiving an indication of a verification of the one or more operation risk(s) via a user interface and / or wherein the one or more operation risk(s) and the configuration and / or operation data are displayed to verify the one or more operation risk(s) are associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities.240150264. The method of any one of claims 1 to 3, further comprising receiving an indication of a verification of the one or more measure(s) via a user interface and / or wherein the one or more measure(s) and the configuration and / or operation data are displayed to verify the one or more measure(s) are associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities.

5. The method of any one of claims 1 to 4, wherein obtaining the measure data includes identifying the one or more measure(s) associated with the one or more operation risk(s) by providing measure determining task instructions to a data-driven model, wherein the data-driven model is configured to follow task instructions, and wherein the measure determining task instructions are related to, in particular include, the one or more identified operation risk(s), and the current configuration and / or operation data.

6. The method of any one of claims 1 to 5, wherein obtaining the operation risk data includes identifying the one or more operation risk(s) by providing risk identification task instructions to a data-driven model, wherein the data-driven model is configured to follow task instructions, and wherein the risk identification task instructions are related to, in particular include, a plurality of operation risks associated with a plurality of chemical production and / or processing facilities and / or the configuration and / or operation data.

7. The method of any one of claims 1 to 6, wherein the measure data and / or operation risk data comprises natural language and / or wherein the data-driven model is trained to process natural language data.

8. The method of any one of claims 1 to 7, further comprising providing a digital identifier suitable for retrieving one or more dataset(s) associated with the at least one measure data and / or operation risk data and wherein obtaining the measure data and / or the operation risk data comprises retrieving the measure data and / or the operation risk data based on the digital identifier.

9. The method of any one of claims 1 to 8, wherein obtaining the request for evaluating the configuration and / or operation of the one or more target chemical production and / or processing facilities is triggered by determining that current configuration and / or operation data associated with the one or more target chemical production and / or processing facilities deviates from target configuration and / or operation data.

10. The method of any one of claims 1 to 9, wherein the current configuration and / or operation data comprises image data related to equipment of the one or more target chemical production and / or processing facilities.1 1. The method of any one of claims 1 to 10, wherein the at least one measure is selected according to a type of the measure associated with an entity affected and / or targeted by the measure.2401502712. The method of any one of claims 1 to 1 1 , wherein obtaining the measure data comprises obtaining an indication of the one or more measure(s) via a user interface, in particular provided by a technical expert associated with the configuration and / or operation of the one or more target chemical production and / or processing facilities.

13. The method of any one of claims 1 to 12, wherein selecting the at least one measure according to the risk score and the risk mitigation score includes obtaining, in particular receiving, a target risk score associated with the chemical production and / or processing facility and selecting the at least one measure according to the risk score, the risk mitigation score and the target risk score, wherein the target risk score is associated with a target occurrence of the corresponding operation risk.

14. An apparatus for mitigating operation risks associated with one or more target chemical production and / or processing facilities, the apparatus comprising: a processor configured for performing any one of the methods according to any one of claims 1 to 13.

15. Use of a measure obtained by any one of claims 1 to 13 for mitigating an operation risk associated with a target chemical production and / or processing facility.

Images