Generating execution protocols for executing standard operating procedures in industrial plants

Abstract

A computer-implemented method (100) for generating and / or extending an execution protocol (4) for at least one standard operating procedure (SOP) (2) in an industrial plant (1), comprising: providing (110) at least one SOP (2) for the plant (1), said SOP (2) comprising a number of steps (2a-2g); providing (120) measurement data (3) acquired during a number of executions of the at least one SOP (2) and indicative of actions taken within the plant (1) that modify a state and / or behavior of the plant (1) or any part thereof in order to execute the SOP (2); determining (130) for each step (2a-2g) of the SOP (2) from the measurement data (3) a subset (3a-3g) of the measurement data (3) indicative of actions taken to execute this particular step (2a-2g) of the SOP; A method comprising a step (140) of aggregating a subset (3a-3g) of the measurement data (3) determined for each step (2a-2g) of the SOP (2) into at least one work instruction (4a-4g) for performing this particular step (2a-2g) of the SOP, wherein the work instruction (4a-4g) is part of a desired protocol (4).

Description

[Technical Field]

[0001] The present invention relates to facilitating the execution of standard operating procedures (SOPs) in industrial plants that are not fully automatable. [Background technology]

[0002] In many safety-critical industrial plants, especially chemical plants, several tasks are performed according to standard operating procedures (SOPs). These SOPs are devised to ensure the safe and reliable operation of the plant. For example, an SOP may specify what to do to start up a plant, shut down a plant, or perform maintenance work on a plant or any part thereof. These and other tasks for which an SOP is required are often not fully automatable but rather require the cooperation of one or more workers who must execute the SOP. SOPs are often part of the authorization or other licensing to operate the plant, and therefore compliance with SOPs is critical.

[0003] EP 1 413 937 A1 discloses a control system for controlling equipment and / or processes based on a finite state machine. The information made available to an operator is enhanced by a graphical representation of the finite state machine on a display. The graphical representation comprises at least two states and at least one allowed transition between these two states. Summary of the Invention

[0004] The object of the present invention is to provide computer-assisted support for the execution of SOPs in industrial plants, even though this execution itself cannot be fully automated.

[0005] This object is achieved by a first computer-implemented method for generating an execution protocol for at least one SOP according to the first independent claim, and by a second computer-implemented method for orchestrating the execution of an SOP using the execution protocol so generated. Further advantageous embodiments are detailed in the respective dependent claims.

[0006] The present invention provides a computer-implemented method for generating and / or extending an execution protocol for at least one standard operating procedure (SOP) in an industrial plant.

[0007] As used herein, the term "protocol" should not be construed as limiting, in the sense that this is the only way to execute an SOP and that everything must be done exactly according to this protocol. Instead, the work instructions within a protocol may allow some flexibility, such as "close all valves leading to a particular vessel" without specifying the order in which the valves should be closed. Also, there may be multiple different ways to execute an SOP, which may be desirable from a reliability standpoint. If a single approach proactively assumes the availability of a piece of equipment and this equipment fails, this is a single point of failure and the SOP cannot continue to execute. However, if there is an alternative approach that achieves the same result using other equipment, there is no longer a single point of failure.

[0008] That is, the term "protocol" should not be construed as limiting, such as "communications protocol." Rather, the meaning of the term "protocol" in the context of the present invention is very similar to that of the term "recipe," i.e., a sequence of operating instructions that allows a particular goal to be achieved. Not coincidentally, the term "execution protocol" is a well-known term in the English language for describing in detail how to inflict the death penalty on a criminal.

[0009] The method begins with at least one SOP for a provided plant. The SOP comprises a plurality of steps. As used herein, the term "step" should not be construed as limiting in the sense that there is only one sequence of steps leading from the start to the end of the SOP. Instead, one or more steps in the SOP may be optional, and there may be a choice between two or more alternative steps at any point during execution of the SOP.

[0010] During the method, measurement data acquired during multiple executions of at least one SOP is provided, the measurement data indicative of actions taken within the plant that modify the state and / or behavior of the plant or any portion thereof to execute the SOP.

[0011] Examples of such measurement data are: Log data indicating the entry of at least one work instruction into the plant's distributed control system to modify the behavior of the plant or any portion thereof. Measurement data delivered by at least one field device that is in direct relation to the industrial process carried out by the plant. · Includes monitoring data indicating the location and / or behavior of at least one worker participating in carrying out the SOP.

[0012] For each step of the SOP, a subset of the measurement data is determined from this measurement data that indicates the actions to be taken to perform this particular step of the SOP, i.e., the measurement data can be divided according to the steps of the SOP to which they pertain, and measurement data that is completely irrelevant to the performance of the SOP can be ignored.

[0013] The subset of measurement data determined for each step of the SOP is aggregated into at least one work instruction for executing this particular step of the SOP. The work instruction is part of the required protocol for executing the SOP as a whole. Specifically, the measurement data may be aggregated through the actions they represent, which may link the measurement data to one or more work instructions. For example, a work instruction may relate to performing one or more actions. Conversely, given one or more actions, the work instruction may be searched for in a predetermined correspondence, e.g., requiring one or more work instructions to be performed in order to perform one or more actions.

[0014] The inventors have found that while SOPs are highly relevant to plant operations, they are often specified at a very abstract level early in the plant's lifecycle. Therefore, they do not necessarily include later adjustments or the knowledge of those with significant experience in this particular plant. Much of this knowledge is difficult to document. Therefore, it often exists only in the minds of the workers who execute the SOPs. Also, the abstract specification of SOPs can be subject to interpretation to some degree.

[0015] For example, if an SOP specifies that a valve should open "slowly," this does not clearly specify whether the opening is to occur at a constant, slow speed, or whether the opening starts very slowly and quickly picks up speed. Also, different operators may have different understandings of the specific speed that "slow" refers to.

[0016] Also, new or revised protocols for executing SOPs include richer details that are especially useful for assisting novices with little experience in the plant. For example, if workers performing a particular step in an SOP are logged as always walking to a certain location in the plant, a protocol for executing an SOP that may have previously only described a piece of equipment by name or in relative terms like "the left-most pump" now includes its specific, unmistakable location on the plant floor. In contrast, the relative term "the left-most pump" could mislead a novices who enter a building through the back door instead of the front door.

[0017] The present invention does not seek to improve the SOP itself. The SOP is taken as given. In the course of the method, a recipe is extracted from the operator's actions to facilitate the execution of the given SOP. That is, the knowledge for implementing the SOP, which may only exist in the operator's mind, can now be documented for use by anyone else.

[0018] SOPs may also reference technical locations that may be difficult to find, such as tags that indicate a specific location. For example, the location indicated on the tag may require a piping and instrumentation diagram to be interpreted by at least a novice worker into a physical location within the plant. However, after multiple visits, an experienced worker will recall the physical location without having to refer to the diagram again.

[0019] Log data may be obtained, for example, from a plant historian who keeps a track record of plant operations and sensor values. For example, log data may include, among other things: Setting new set points for low-level controllers in the plant, Opening or closing at least one valve in the plant; Start or stop at least one piece of equipment in the plant; and Commanding any other actuators to modify the state or behavior of the plant or any part thereof may indicate one or more of:

[0020] The monitoring data indicating the location and / or behavior of at least one worker participating in carrying out the SOP may be, e.g. At least one video stream showing at least one worker participating in executing the SOP and / or captured by a camera worn by this worker; data indicating the gaze direction of at least one worker participating in the execution of the SOP; A radio or audio recording of the voice of at least one worker participating in carrying out the SOP; and Interactions, such as cursor movement and menu selection, between at least one operator participating in the execution of the SOP and the human-machine interface of the plant's distributed control system may comprise one or more of:

[0021] For example, a factory floor can be monitored by video surveillance to track where workers are going within the plant. The location of the workers can also be obtained from video streams captured by cameras worn by the workers. The latter, as well as the worker's gaze direction, can also be used to monitor what the worker is actually paying attention to.

[0022] A radio or audio recording of the worker's voice can yield information about what communication between this worker and others in the plant is necessary to carry out the steps of the SOP. For example, if the SOP for maintenance of a piece of equipment mandates that this equipment be disconnected from the main power source and protected from reconnection, this may require communication with a central control room by two-way radio to have the power disconnected and the respective switches tagged and locked out.

[0023] Interaction between the worker and the human-machine interface may yield information about what information the worker is searching for within this interface. For example, an SOP may specify that the internal temperature and / or pressure must be checked before a container is opened. Interaction with the human-machine interface may augment the specifications in the SOP with information if the needed information can be found.

[0024] Determining the respective actions "for performing each specific step of the SOP" means that, given each step of the SOP, it is determined which manual actions are to be performed for this specific step. Thus, the manual actions and measurement data indicative of these manual actions are grouped according to the given SOP step. Thus, the process is still governed by the original SOP, which has not been modified or overwritten.

[0025] In a particularly advantageous embodiment, determining the subset of measurement data for a step of the SOP comprises: The time slot in which the step in the SOP was performed; the relevant equipment for carrying out the steps of the SOP, Locations within the plant that are relevant for carrying out the steps in the SOP; Workers who are assigned roles to participate in carrying out the steps of the SOP, and Graphical user interface components that are highly relevant to execution That is, a "fuzzy alignment" process may be performed to cluster measurement data that is relevant to a particular SOP step, thereby filtering out data that is not relevant to the performance of the SOP, and allowing the remaining data to be mapped to the SOP steps.

[0026] The equipment involved in performing the steps of the SOP can be extracted from the SOP description, for example, by appropriate heuristics. Similarly, the roles of the workers required to participate in performing the steps of the SOP can also be extracted from the SOP description. The time slots in which the steps of the SOP were performed can be extracted from the measurement data itself in combination with the SOP description. For example, the completion of some steps of the SOP can appear in some signature in the measurement data, such as a record of a valve closing or a flow dropping to zero.

[0027] Each single filtering criterion applied to the measurement data can be considered a "weak learner" that provides some awareness, but no clear judgment, of the relevance of a record of the measurement data for a particular step of the SOP. However, a combination of multiple such "weak learners" can provide a much more accurate assessment of whether a record of the measurement data is relevant for performing a particular step of the SOP. Thus, in a particularly advantageous embodiment, the measurement data is filtered according to multiple criteria. A relevance score is assigned to each record of the measurement data. This relevance score increases with the number of criteria that this particular record of the measurement data satisfies. The record of the measurement data can then be included in a subset for a particular step of the SOP in response to the relevance score exceeding a predetermined threshold.

[0028] In a further advantageous embodiment, the method further comprises determining, from the measurement data, from the previously determined subset, and / or from the previously determined work instructions, at least one directed sequence of actions that results in the execution of the complete SOP as the desired protocol. This means that the method can not only fill in more details of the execution protocol for the SOP, but also reveal one or more ways to consider the SOP as a whole. The SOP itself may not be clear about the order in which certain actions need to be performed. In particular, the method may link a single graph of actions derived from the measurements to the SOP, but there may be multiple ways to consider this graph.

[0029] For example, an SOP may only require powering up a machine, which may be controlled by a separate control unit. Here, there are two approaches to powering up the machine: power up the control unit first, then the machine, or power up the machine first, then the control unit. At first glance, there are reasons for both approaches. Powering up the control unit first ensures that the machine does not have power without being under the control of the control unit. Powering up the machine first ensures that the control unit is not confused because the machine is “missing” when it is starting up. From measurement data, the method can learn which of these two reasons seems more plausible in the context of this particular plant. Also, the work instructions and / or more detailed support communicated to the operator can be somewhat dynamic as a result of examining the action graph differently. For example, whether the operator powers up the machine or the control unit first, the operator may always receive dynamic support to operate the equipment he or she chose to start up first.

[0030] There are many more examples of how an SOP can be implemented for multiple different paths. Having multiple such paths may even be desirable for safety reasons. For example, releasing pressure inside a vessel may normally be performed by an automatically controlled valve, but if this valve sticks or there is a power failure, it may be necessary to release the pressure using a manually operated valve.

[0031] Thus, in a further advantageous embodiment, multiple sequences of actions that result in the execution of a complete SOP are combined into a directed graph of these actions. In this way, the method can learn all available alternatives for moving forward in the SOP, including alternatives that were not foreseen when establishing the SOP. For example, if a piece of equipment breaks down, plant personnel may find creative ways to replace the lost functionality by repurposing other equipment.

[0032] For example, if it is not possible to remove a large amount of acid from a reaction vessel because a subsequent vessel in which the acid is neutralized is leaking, a different port on the reaction vessel may be diverted to transfer the acid to another vessel elsewhere in the plant so that the acid can be neutralized there and safely released.

[0033] Preferably, in this directed graph, each edge connecting a first action to a subsequent second action is assigned a probability that the second action will be performed given the first action, based on the provided measurement data, which immediately reveals what are preferred and secondary approaches to proceed in the SOP.

[0034] Therefore, in a further advantageous embodiment, for each action connected to two or more possible next actions, a path through the graph that selects the next action with the highest probability is determined as the execution protocol for the SOP. In this way, if different techniques for performing a step of the SOP have previously been used by different workers, a single standard technique may be created that should be used unless there is a compelling reason for deviation (such as a malfunction). Having a single standard technique reduces the possibility of error, thereby increasing the probability that the SOP will be performed correctly. For example, if a first worker performs an SOP and the single technique observes that a second worker is performing a particular step differently, the first worker may adopt this behavior in the context of the path he or she is following, even though it is incorrect. That is, multiple different execution protocols may be adopted by different workers, resulting in a mixture of different execution protocols that violate the SOP.

[0035] The present invention also provides a computer-implemented method for orchestrating the execution of at least one standard operating procedure (SOP) in an industrial plant.

[0036] In the course of this method, at least one execution protocol for the SOP generated and / or extended by the aforementioned method is provided. This generation and / or extension may be performed by the same business entity that is organizing the execution of the SOP, or may be performed at least in part by a different business entity. For example, a company that owns a plant where the SOP will be implemented may enlist the help of another company that offers the generation of execution protocols from measurement data as a service.

[0037] Measurement data is obtained that indicates at least one activity being performed by at least one operator who will participate in executing the SOP. For example, the measurement data may indicate the location or gaze direction of the operator, or what the operator is currently looking at. The measurement data may also relate to, for example, the state of the plant or any part thereof. For example, the measurement data may indicate if the operator has already opened or closed a valve that is scheduled to be opened or closed.

[0038] At least one work instruction to be performed by the worker is selected from the execution protocol based at least in part on the measurement data. The work instruction to perform the at least one action is communicated to the worker. This communication may be performed in any suitable manner. For example, this information may be overlaid on an augmented reality display used by the worker. For example, the worker may be guided to a location where the action will be performed. Equipment that the worker should handle may then be highlighted in the augmented reality display. When the worker is interacting with the plant via a human-machine interface of the distributed control system, In a particularly advantageous embodiment, during the process of selecting a work order, it is determined whether the worker has already performed the work order and / or is already at the location where this work order is to be performed. If so, the work order and / or instructions for guiding the worker to the location where this work order is to be performed are suppressed. In this way, the worker is not overloaded with unnecessary information that can be safely ignored, which could lead the worker to ignore other, more safety-related information.

[0039] Similarly, in a further advantageous embodiment, the skill level of the worker is determined based on current and / or past measurement data. The level of detail of the work instructions is based on this skill level. For example, a worker already familiar with the plant may only need work instructions to get to equipment using a specific name or other label, while a novice worker may need step-by-step work instructions on how to find the equipment. For example, if a worker always goes to the right place when instructed to go to a particular piece of equipment, the level of detail of the work instructions may gradually decrease.

[0040] The computer implementation of the above methods implies that the methods may be embodied in a computer program. Accordingly, the present invention also provides a computer program having machine-readable instructions that, when executed by one or more computers, cause the one or more computers to perform one of the above methods. The present invention also provides a non-transitory machine-readable storage medium having the computer program and / or a download product. The download product is a product that can be sold in an online shop for immediate realization by download. The present invention also provides one or more computers having the computer program and / or having the non-transitory machine-readable storage medium and / or download product.

[0041] In the following, the invention is illustrated by means of figures, without intending to limit the scope of the invention. [Brief explanation of the drawings]

[0042] [Figure 1] 1 is a diagram of an exemplary embodiment of a method 100 for generating and / or extending an execution protocol 4 for SOP2. [Figure 2] 1 is a diagram of an exemplary execution protocol 4 generated by method 100. FIG. [Figure 3] 2 is a diagram of an exemplary embodiment of a method 200 for orchestrating the execution of SOP2. DETAILED DESCRIPTION OF THE INVENTION

[0043] FIG. 1 is a schematic flow chart of an embodiment of a method 100 for generating and / or extending an execution protocol 4 for SOP2.

[0044] In step 110, at least one SOP 2 is provided, comprising a plurality of steps 2a-2g. In step 120, measurement data 3 is provided, the measurement data 3 having been acquired during a plurality of executions of the at least one SOP 2. The measurement data 3 is indicative of actions taken within the plant 1 that modify the state and / or behavior of the plant 1 or any part thereof to implement the SOP 2.

[0045] In step 130, for each step 2a-2g of the SOP 2, a subset 3a-3g of the measurement data 3 is determined that is indicative of the actions to be performed to perform this particular step 2a-2g of the SOP.

[0046] According to block 131, this determining 130 may in particular comprise "fuzzy filtering" from the measurement data according to one or more criteria, each of which may be a "weak learner". In particular, according to block 131a, the measurement data 3 may be filtered according to a plurality of criteria. According to block 131b, each record of the measurement data 3 may be assigned a relevance score that increases with the number of criteria that this particular record of the measurement data 3 fulfills. In response to this relevance score exceeding a predetermined threshold, according to block 131c, the record of the measurement data 3 may be included in a subset 3a-3g.

[0047] In step 140, the subset 3a-3g of measurement data 3 determined for each step 2a-2g of the SOP 2 is aggregated into at least one work instruction 4a-4g for performing this particular step 2a-2g of the SOP. This work instruction 4a-4g is part of the desired protocol 4.

[0048] In step 150, a directed sequence of actions resulting in the execution of the complete SOP 2 may be determined from the measurement data 3, from the subsets 3a-3g, and / or from the work instructions 4a-4g as the desired protocol 4. These actions may correspond to work instructions 4a-5g, although one or more of the work instructions 4a-4g may also be decomposed into more granular sub-actions.

[0049] According to block 151, multiple sequences of actions that result in the execution of a complete SOP2 may be combined into a directed graph of these actions. Specifically, according to block 151a, each edge connecting a first action to a subsequent second action may be assigned a probability that the second action will be executed given the first action. According to block 151b, for each action connected to two or more possible next actions, a path through the graph that selects the next action with the higher probability may be determined as execution protocol 4 for SOP2.

[0050] Figure 2 is an example of an execution protocol 4 for SOP2. Execution protocol 4 is in the form of a directed graph and comprises work instructions 4a-4g corresponding to steps 2a-2g of SOP2. After the second step 4b, there are two options. The first option is to perform work instruction 4c. The second option is to perform a combination of work instructions 4d and 4e. Both options reach the same result, which is to put plant 1 in a state where the next work instruction 4f can be performed, and then finally work instruction 4g.

[0051] For example, work instruction 4c may refer to a simple approach to achieving a required result, while the combination of work instructions 4d and 4e may refer to an indirect approach to achieving the same result. Thus, as shown in Figure 2, according to measurement data 3, the direct approach is followed 80% of the time, while the indirect approach is followed 20% of the time. The indirect approach may be useful, for example, as a backup if the equipment required for the direct approach is not available.

[0052] FIG. 3 is a schematic flow chart of an exemplary embodiment of a method 200 for orchestrating the execution of at least one SOP2.

[0053] In step 210, at least one execution protocol 4 is provided having work instructions 4a-4g generated and / or extended by the method 100 described above. In step 220, measurement data 3 is obtained. The measurement data 3 is indicative of at least one activity being performed by at least one worker 5. The worker 5 will participate in executing the SOP 2.

[0054] In step 230, at least one work instruction 4a-5g to be performed by the worker 5 is selected from the execution protocol 4 based at least in part on the measurement data 3. Herein, according to block 231, it may be determined whether the worker 5 has already performed the work instruction 4a-4g and / or is already at the location where this work instruction 4a-4g is to be performed. If so (true value 1), according to block 232, the work instruction 4a-4g and / or instructions for guiding the worker 5 to the location where this work instruction 4a-4g is to be performed may be suppressed.

[0055] In step 240, work instructions 4a-4g are communicated to a worker 5. Herein, according to block 241, the skill level of the worker 5 may be determined based on current and / or past measurement data 3. According to block 242, the level of detail of the work instructions may be adjusted based on the skill level. [Explanation of symbols]

[0056] 1. Industrial plants 2 Standard Operating Procedures (SOPs) Steps 2a to 2g of SOP2 3. Measurement data 3a-3g: Subset of measurement data corresponding to steps 2a-2g 4. Execution Protocol for SOP2 4a~4g Work instructions for Execution Protocol 4 5. Workers 100 Method for generating and / or extending an execution protocol 4 110 Provide SOP2 120 Provide measurement data 3 130 Determine subsets 3a-3g of measurement data 3 for steps 2a-2g. 131 Fuzzy filtering of measurement data 3. 131a Filtering according to multiple criteria 131b Assign relevance to measurement data 3 131c Consider measurement data3 based on relevance 140 Aggregate subsets 3a to 3g into work instructions 4a to 4g 150 Determine a directed sequence of actions 151 Create a directed graph of actions 151a Assigning Probabilities to Edges 151b Determine the path with the greatest probability 200 Methods for organizing the implementation of SOP2 210 Provide Execution Protocol 4 220 Obtain measurement data 3 230 Select work instructions 4a to 4g 231 Determine whether work orders 4a-4g have been performed or arrived at the location 232 Suppress work instructions 4a to 4g 240 Communicate work instructions 4a to 4g to worker 5 241 Determine the skill level of worker 5 242 Adjusting Detail Based on Skill Level The inventions described in the claims of the present application as originally filed are set forth below. [C1] A computer-implemented method (100) for generating and / or extending an execution protocol (4) for at least one standard operating procedure (SOP) (2) in an industrial plant (1), comprising: A step (110) of providing at least one SOP (2) for the plant (1), the SOP (2) comprising a plurality of steps (2a to 2g); providing (120) measurement data (3) acquired during multiple executions of the at least one SOP (2) and indicative of actions taken within the plant (1) that modify the state and / or behavior of the plant (1) or any part thereof to execute the SOP (2); a step (130) of determining from the measurement data (3) for each step (2a-2g) of the SOP (2) a subset (3a-3g) of the measurement data (3) indicative of an action to be taken to perform this particular step (2a-2g) of the SOP; and a step (140) of aggregating the subset (3a-3g) of the measurement data (3) determined for each step (2a-2g) of the SOP (2) into at least one work instruction (4a-4g) for performing this particular step (2a-2g) of the SOP, wherein this work instruction (4a-4g) is part of the desired protocol (4). A method for providing [C2] The measurement data (3) is log data indicative of the input of the at least one work instruction for modifying the behavior of the plant (1) or any part thereof into a distributed control system of the plant; Measurement data (3) delivered by at least one field device in direct relation to the industrial process carried out by said plant (1), and Surveillance data indicating the location and / or behavior of at least one worker (5) participating in implementing said SOP (2). The method (100) according to C1, comprising one or more of: [C3] The log data may, in particular, setting new set points for low-level controllers in the plant (1); opening or closing at least one valve in the plant (1); Starting or stopping at least one device in the plant (1); and command of any other actuators that modify the state or behavior of said plant (1) or any part thereof. The method (100) according to C2, wherein the method (100) shows one or more of the following: [C4] The monitoring data includes: at least one video stream showing at least one worker (5) participating in executing the SOP (2) and / or captured by a camera worn by this worker (5); data indicating the gaze direction of at least one worker (5) participating in executing the SOP (2); a radio or audio recording of the voice of at least one worker (5) participating in implementing said SOP (2); and Interactions, such as cursor movement and menu selection, between at least one operator (5) participating in executing the SOP (2) and a human-machine interface of the distributed control system of the plant (1). The method (100) according to any one of C2 to C3, comprising one or more of: [C5] The step (130) of determining a subset (3a to 3g) of the measurement data (3) relating to steps (2a to 2g) of the SOP (2) particularly determines from the measurement data (3): the time slots in which the steps (2a to 2g) of the SOP (2) were performed; Equipment relevant for carrying out the steps (2a to 2g) of the SOP (2), Locations within the plant (1) relevant for carrying out the steps (2a to 2g) of the SOP (2), A worker (5) who is assigned a role to participate in performing the steps (2a to 2g) of the SOP (2); and Graphical user interface components highly relevant to said implementation The method (100) according to any one of C1 to C4, comprising filtering (131) a portion relating to one or more of: [C6] filtering (131a) said measured data (3) according to a plurality of criteria; and assigning (131b) a relevance score to each record of the measured data (3), wherein the relevance score increases with the number of criteria that this particular record of the measured data (3) satisfies. including (131c) a record of the measurement data (3) within the subset (3a-3g) in response to the relevance score exceeding a predetermined threshold; and The method (100) according to C5, further comprising: [C7] The method (100) according to any one of C1 to C6, further comprising determining (150) from the measurement data (3), from the subsets (3a to 3g) and / or from the work instructions (4a to 4g) at least one directed sequence of actions that results in the execution of a complete SOP (2) as the required protocol (4). [C8] The method (100) of C7 further comprising combining (151) multiple sequences of actions that result in the execution of a complete SOP (2) into a directed graph of these actions. [C9] The method (100) of C8, further comprising assigning (151a) to each edge connecting a first action and a second action a probability that the second action will be performed given the first action based on the provided measurement data (3). [C10] The method (100) of C9 further comprises, as an execution protocol (4) for the SOP (2), determining (151b) a path through the graph that selects, for each action connected to two or more possible next actions, a next action having a higher probability. [C11] A computer-implemented method (200) for orchestrating the execution of at least one standard operating procedure (SOP) (2) in an industrial plant (1), comprising: Providing (210) at least one execution protocol (4) for the SOP (2) generated and / or extended by the method (100) according to any one of C1 to C10; A step (220) of obtaining measurement data (3) indicative of at least one activity being performed by at least one worker (5) who will participate in executing the SOP (2); and a step (230) of selecting, at least in part based on the measurement data (3), at least one work instruction (4a-4g) from the execution protocol (4) to be performed by the worker (5). a step (240) of communicating the work instructions (4a to 4g) to the worker (5); A method for providing [C12] The step of selecting at least one work instruction comprises: determining (231) whether the worker (5) has already performed a work instruction (4a-4g) and / or is already at the location where the work instruction (4a-4g) is to be performed; If so, suppressing (232) this work instruction (4a-4g) and / or instructions for guiding to the location where this work instruction (4a-4g) is to be performed. [C13] The step (240) of communicating the work instructions (4a to 4g) includes: determining (241) a skill level of said worker (5) based on current and / or historical measurement data (3); adjusting (242) the level of detail of the work instructions (4a to 4g) based on this skill level; The method (200) according to C11 or 12, comprising: [C14] A computer program comprising machine-readable instructions that, when executed by one or more computers, cause the one or more computers to perform a method (100, 200) according to any one of C1 to C13. [C15] A non-transitory machine-readable storage medium and / or download product having a computer program according to C14. [C16] One or more computers having a computer program as described in C14 and / or a non-transitory storage medium and / or download product as described in C15.

Claims

1. A computer-implemented method (100) for generating and / or extending an execution protocol (4) for at least one standard operating procedure (SOP) (2) in an industrial plant (1), comprising: a step (110) of providing at least one SOP (2) for the industrial plant (1), the at least one SOP (2) comprising a plurality of steps (2a to 2g); providing (120) measurement data (3) acquired during the execution of the steps (2a-2g) of the at least one SOP (2) and indicative of actions taken within the industrial plant (1) that modify the state and / or behavior of the industrial plant (1) or any part thereof in order to execute the at least one SOP (2); a step (130) of determining, for each step (2a-2g) of the at least one SOP (2), a subset (3a-3g) of the measurement data (3) indicative of actions to be performed to execute this particular step (2a-2g) of the at least one SOP (2), the step of determining the subset (3a-3g) comprising dividing the measurement data (3) according to the step of the SOP (2) to which the measurement data (3) relates; aggregating (140) the subset (3a-3g) of the measurement data (3) determined for each step (2a-2g) of the at least one SOP (2) into at least one work instruction (4a-4g) for executing this particular step (2a-2g) of the at least one SOP (2), wherein this work instruction (4a-4g) is part of the desired execution protocol (4); A method for providing

2. The measurement data (3) is log data indicative of the input of the at least one work instruction for modifying a state or behavior of the industrial plant (1) or any part thereof into a distributed control system of the industrial plant (1); Measurement data (3) delivered by at least one field device in direct relation to the industrial process carried out by said industrial plant (1), and monitoring data indicative of the location and / or behavior of at least one worker (5) participating in the execution of said at least one SOP (2); The method (100) of claim 1, comprising one or more of:

3. The log data is Setting new set points for low level controllers in the industrial plant (1); Opening or closing at least one valve in the industrial plant (1), Starting or stopping at least one device in the industrial plant (1), and command of any other actuators that modify the state or behavior of said industrial plant (1) or any part thereof. The method (100) of claim 2, wherein the method (100) exhibits one or more of the following:

4. The monitoring data includes: at least one video stream showing at least one worker (5) participating in the execution of said at least one SOP (2) and / or captured by a camera worn by this worker (5); data indicating the gaze direction of at least one worker (5) participating in the execution of said at least one SOP (2); a radio or audio recording of the voice of at least one worker (5) participating in executing said at least one SOP (2); and Interactions, such as cursor movements and menu selections, between at least one operator (5) participating in the execution of said at least one SOP (2) and a human-machine interface of said distributed control system of said industrial plant (1). The method (100) of claim 2, comprising one or more of:

5. The step (130) of determining a subset (3a-3g) of the measurement data (3) relating to steps (2a-2g) of the at least one SOP (2) includes determining from the measurement data (3): the time slot in which the steps (2a to 2g) of the at least one SOP (2) were performed; relevant equipment for carrying out said steps (2a-2g) of said at least one SOP (2); Locations within the industrial plant (1) that are relevant for carrying out the steps (2a-2g) of the at least one SOP (2), A worker (5) who is assigned a role to participate in performing said steps (2a-2g) of said at least one SOP (2); and Graphical user interface components highly relevant to the execution protocol (4) The method (100) of any one of claims 1 to 4, comprising filtering (131) a portion relating to one or more of:

6. filtering (131a) the measurement data (3) according to a plurality of criteria; assigning (131b) a relevance score to each record of said measured data (3), wherein said relevance score increases with the number of criteria that this particular record of said measured data (3) satisfies; including (131c) a record of said measurement data (3) within said subset (3a-3g) in response to said relevance score exceeding a predetermined threshold; The method (100) of claim 5 further comprising:

7. The method (100) according to any one of claims 1 to 6, further comprising determining (150) from the measurement data (3), from the subsets (3a-3g) and / or from the work instructions (4a-4g) at least one directed sequence of actions that results in the execution of a complete SOP (2) as the execution protocol (4) sought.

8. 8. The method (100) of claim 7, further comprising combining (151) multiple sequences of actions that result in the execution of the complete SOP (2) into a directed graph of these actions.

9. 9. The method (100) of claim 8, further comprising assigning (151 a) to each edge connecting a first action and a second action a probability that the second action will be performed given the first action based on the provided measurement data (3).

10. 10. The method of claim 9, further comprising: determining, as an execution protocol for the complete SOP, a path through the directed graph that selects, for each action connected to two or more possible next actions, a next action having a higher probability.

11. aggregating the subsets (3a-3g) of the measurement data (3) determined for each step (2a-2g) of the SOP (2), aggregating said measurement data (3) through the actions they represent; and / or The method of claim 1 , comprising searching within a predetermined correspondence which one or more work instructions need to be executed to perform one or more actions.

12. The method of claim 1, wherein the step (130) of determining, for each step (2a to 2g) of the at least one SOP (2), from the measurement data (3), a subset (3a to 3g) of the measurement data (3) indicative of actions to be performed to execute this particular step (2a to 2g) of the at least one SOP (2) means that, given each step (2a to 2g) of the SOP (2), it is determined which manual actions are to be performed for this particular step, and the manual actions and the measurement data (3) indicative of the manual actions are grouped according to each step (2a to 2g) of the given SOP (2).

13. A computer-implemented method (200) for orchestrating the execution of at least one standard operating procedure (SOP) (2) in an industrial plant (1), comprising: providing (210) at least one execution protocol (4) for said at least one SOP (2) generated and / or extended by the method (100) of any one of claims 1 to 12; obtaining (220) measurement data (3) indicative of at least one activity being performed by at least one worker (5) who will participate in executing said at least one SOP (2); selecting (230) at least one work instruction (4a-4g) from the execution protocol (4) to be performed by the at least one worker (5) based at least in part on the measurement data (3); a step (240) of communicating the work instructions (4a to 4g) to the at least one worker (5); A method for providing

14. The step of selecting at least one work instruction comprises: determining (231) whether said at least one worker (5) has already performed said work order (4a-4g) and / or is already at the location where said work order (4a-4g) is to be performed; If so, suppressing (232) this work instruction (4a-4g) and / or instructions for guiding to the location where this work instruction (4a-4g) is to be performed; 14. The method (200) of claim 13, comprising:

15. The step (240) of communicating the work instructions (4a to 4g) includes: determining (241) a skill level of said at least one worker (5) based on current and / or historical measurement data (3); adjusting (242) the level of detail of the work instructions (4a-4g) based on the skill level; 15. The method (200) of claim 13 or 14, comprising:

16. A computer program comprising machine-readable instructions which, when executed by one or more computers, cause the one or more computers to perform the method (100, 200) of any one of claims 1 to 15.

17. 17. A non-transitory machine-readable storage medium having the computer program of claim 16.

18. 18. One or more computers comprising a computer program according to claim 16 and / or a non-transitory machine-readable storage medium according to claim 17.

Images