Method of fulfilling a maintenance task by a worker, and data processing apparatus

Abstract

A method of fulfilling a maintenance task by a worker is described, including receiving a maintenance task indicative of a maintenance procedure to be performed; generating a worker dataset comprising worker data of at least one worker, the worker data comprising worker skill data indicative of at least one of a plurality of worker skills; generating a procedure dataset comprising procedure data for maintenance procedures, the procedure data being indicative of at least one worker skill requirement defining a worker skill required for performing a respective maintenance procedure from the supported maintenance procedures; determining a competence score of the at least one worker, the competence score being determined based on the worker skill requirement of the maintenance procedure corresponding to the maintenance task, and the worker skill data of the worker; selecting at least one worker based on the competence score; and performing the maintenance procedure by the selected worker.

Description

[0001] Method of fulfilling a maintenance task by a worker, and data processing apparatus

[0002] Aspects of the invention relate to fulfilling a maintenance task by a worker, in particular selecting a worker to perform a maintenance procedure. Further aspects relate to receiving a maintenance task, determining a competence score of the worker, and selecting the worker based on the competence score. Aspects of the invention further relate to a data processing apparatus configured for selecting a worker based on a maintenance task and a worker competence score.

[0003] Technical background:

[0004] Modern sites, such as industrial installations, mines, pumps, power-, transport-, telecommunications- and / or supply-infrastructure, often contain assets, such as machines or other types of equipment, which may require maintenance to be performed by a worker.

[0005] Such modern sites may operate (semi-)autonomously, which can result in less workers being routinely tied to a particular site, asset, or task. In some cases, a modern workforce may constitute a global pool of workers each having a worker-specific skill set. When selecting workers from the pool of workers, there is a risk that the selected worker is unfamiliar with the task, the site and / or the asset.

[0006] Selecting the right worker for a particular task, such as repair, maintenance, installation, retrofitting or configuration of an asset is often a core task of maintenance planning and service execution. Matching worker skills and work demands is often beneficial. For example, assigning cleared and capable workers to a maintenance task may reduce the risk of downtime, accidents and / or task failure, increase efficiency and reduce the cost of performing the maintenance task.

[0007] Worker skill is often a result of work experience gained through having fulfilled maintenance tasks in the past, and worker training, such as training courses, in which specific skills are taught to the worker. Unfavorably, workers are often unreachable during training, and therefore not available for fulfilling maintenance tasks. It may be beneficial to determine training requirements of the worker, based on the worker skill, a skill requirement of the maintenance task, and / or even a (potential) training effect of the worker fulfilling the maintenance task.

[0008] In light of these factors, efficiently and accurately selecting a worker for a maintenance task and / or training can be challenging. The systems, apparatus and methods described herein solve the above-stated problem at least in part. Summary of the invention:

[0009] The invention is set out in the appended set of claims.

[0010] According to an aspect, a method of fulfilling a maintenance task by a worker is described. The method includes receiving a maintenance task indicative of a maintenance procedure to be performed; generating a worker dataset comprising worker data of at least one worker, the worker data comprising worker skill data indicative of at least one of a plurality of worker skills; generating a procedure dataset comprising procedure data for maintenance procedures, the procedure data being indicative of at least one worker skill requirement defining a worker skill required for performing a respective maintenance procedure from the supported maintenance procedures; determining a competence score of the at least one worker, the competence score being determined based on the worker skill requirement of the maintenance procedure corresponding to the maintenance task, and the worker skill data of the worker; selecting at least one worker based on the competence score; and performing the maintenance procedure by the selected worker.

[0011] According to an aspect, a data processing apparatus is described, including a database with worker data of at least one worker, the worker data comprising worker skill data indicative of at least one of a plurality of worker skills; a procedure database with procedure data for supported maintenance procedures, the procedure data being indicative of at least one worker skill requirement defining a worker skill required for performing a respective maintenance procedure from the supported maintenance procedures; an interface configured for receiving a maintenance task indicative of a maintenance procedure to be performed; and a processor configured for determining a competence score of the at least one worker, the competence score being determined based on the worker skill requirement of the maintenance procedure corresponding to the maintenance task, and the worker skill data of the worker. The data processing apparatus is configured for selecting at least one worker based on the competence score; and generating an output indicative of the selected worker.

[0012] According to an aspect, a worker is described. The worker may be a maintenance worker, service technician, equipment mechanic, maintenance and / or repair specialist, maintenance engineer or other type of professional. The present invention is not limited to any specific type of technical field, accordingly, workers may also include other types of maintenance personnel associated with specialized assets, such as software specialists, food technicians, petrochemical engineers, pharmaceutical technicians, plumbers, electricians, and / or mobile equipment technicians. The worker may be a worker within a pool of available workers, such as workers that are “on call” for traveling within an industrial site, industrial complex, or even to remote sites in case maintenance and / or repair is required.

[0013] According to an aspect, maintenance is described. Maintenance, in the context of this disclosure, is to be understood as an operation or set of operations in which a worker interacts with an asset to achieve a goal. Maintenance is not limited to maintaining an existing asset, and may further include providing, provisioning or installing new or additional assets, removing and / or replacing assets, (re-)configuring an asset, or even operating an asset. For example, and not limited thereto, maintenance may include performing routine maintenance, such as lubricating a machine or exchanging consumable items of the machine. Maintenance may further include repairing a broken machine, e.g. by replacing parts of the machine, or even the whole machine. Maintenance may further include installing new machines, e.g. in place of decommissioned machines. Maintenance may include configuring a controller of a machine or set of machines. Maintenance may include maintaining infrastructure related to the operation of an asset.

[0014] According to an aspect, a maintenance procedure is described. A maintenance procedure may be understood as an operation or (limited) set of operations to be performed by a worker on a single component or a defined set of e.g. interoperating components. For example, a maintenance procedure may be a procedure as defined in a service, operation and / or maintenance manual. Maintenance of an asset may include performing one or several maintenance procedures. For example, an asset may include several components, and the maintenance of each component may include performing one or more maintenance procedures. For example, and not limited thereto, an asset, such as a pumping station, may include a motor and a pump, and maintaining the pumping station may include performing maintenance procedures directed at the motor, and maintenance procedures directed at the pump.

[0015] According to an aspect, a maintenance task is described. The maintenance task may include data indicative of one or more assets, and the maintenance procedures to be performed on the asset(s). The maintenance task may include a problem description and / or diagnostic data. The maintenance task may include further data, such as a location of the asset, an (expected) environment of the asset, such as a geographic location of the asset, weather data, information indicative of the local safety and / or political climate, and other data that may relate to and / or be indicative of worker skills required for performing the maintenance task. For example, and not limited thereto, the maintenance task may be generated by and / or obtained from a maintenance management system, such as a computerized maintenance management system (CM MS). The maintenance task may be indicative of at least one maintenance procedure to be performed, and may further be indicative of a plurality of maintenance procedures to be performed. Each maintenance procedure of the plurality of maintenance procedures may have different worker skill requirements.

[0016] Beneficially, the methods and apparatus described herein allow the efficient selection of competent workers for fulfilling a maintenance task. Furthermore, embodiments of the present invention may beneficially increase a skill level of a worker and / or a pool of workers, and may beneficially provide deep skill management, situation-specific training and training-aware maintenance.

[0017] Further advantages, features, aspects and details that can be combined with embodiments described herein are evident from the dependent claims, the description and the drawings.

[0018] Brief description of the Figures:

[0019] The details will be described in the following with reference to the figures, wherein

[0020] Fig. 1 is a schematic flowchart of a method of fulfilling a maintenance task according to embodiments;

[0021] Fig. 2 is a schematic visualization of a worker competence according to embodiments;

[0022] Fig. 3 schematically shows a data processing apparatus according to embodiments.

[0023] Detailed description of the Figures and of embodiments:

[0024] Reference will now be made in detail to the various embodiments, one or more examples of which are illustrated in each figure. Each example is provided by way of explanation and is not meant as a limitation. For example, features illustrated or described as part of one embodiment can be used on or in conjunction with any other embodiment to yield yet a further embodiment. It is intended that the present disclosure includes such modifications and variations.

[0025] Within the following description of the drawings, the same reference numbers refer to the same or to similar components. Generally, only the differences with respect to the individual embodiments are described. Unless specified otherwise, the description of a part or aspect in one embodiment applies to a corresponding part or aspect in another embodiment as well.

[0026] Referring now to Fig. 1 , a method 100 of fulfilling a maintenance task by a worker is described. As will become clear by the following description, the method 100 does not necessarily need to be performed in the order shown in Fig. 1 , and some of the operations may be performed independently of other operations.

[0027] The method 100 includes generating a procedure dataset comprising procedure data for maintenance procedures. The procedure data is indicative of worker skill requirements for performing one or more supported maintenance procedures. In operation 110, supported maintenance procedures are defined. For example, operation 110 may include retrieving, defining or generating data indicative of one or more maintenance procedures. For example, operation 110 may include retrieving supported maintenance procedures from a general procedure database, and / or generating a database of supported maintenance procedures. For example, operation 110 may include generating a database of supported maintenance procedures according to technical information, such as maintenance and / or repair manuals, repair logs, or other types of documentation. Operation 110 may include generating a dataset comprising procedure data defining the steps to be carried out by a worker to fulfil a maintenance procedure, such as an indication of the type of asset to be maintained in the maintenance procedure, and / or an indication of which tools and / or procedures to be used for the maintenance procedure by the worker.

[0028] In operation 112, worker skill requirements are defined for the maintenance procedures defined in operation 110. For example, the procedure data may be analyzed, e.g. based on the asset type, the tools, and / or the procedures to be used in the maintenance procedure, and a required worker skill may be determined based on the analysis. For example, a maintenance procedure to be carried out on a specific asset type, such as a e.g. a motor, a pump, a controller, may be indicative of a worker skill requirement defining that the worker should be experienced and / or competent to perform work on the specific asset type. For example, a maintenance procedure requiring the use of a specific toolset, such as specialized tools, and / or a specific procedure, such as e.g. a specific calibration or other type of maintenance procedure, may be indicative of a worker skill requirement defining that the worker should be experienced and / or competent to use the specific toolset and / or perform the specific procedure. It should be noted that a maintenance procedure and the corresponding worker skill requirement does not necessarily need to be tied to a specific asset and / or toolset. For example, a maintenance procedure may be defined by the type of environment a maintenance procedure is to be carried out in, which may require specific worker skills. For example, and not limited thereto, a maintenance procedure may require the worker to have a worker skill indicating that the worker is skilled in operating e.g. under extreme conditions, such as extreme cold and winds to be expected e.g. in the arctic tundra. Operations 110, 112, may result in the generation of a procedure dataset containing procedure data of a plurality of maintenance procedures, each maintenance procedure defining one or more worker skill requirements for performing the maintenance procedure.

[0029] The method 100 includes, in operation 120, generating a worker dataset comprising worker data of at least one worker. Generating the worker dataset may include generating, modifying, maintaining, updating and / or appending a dataset including worker data for some workers, or even all workers in a workforce. Operation 120 may include maintaining the dataset, e.g. to update the worker data based on the work or training a worker has experienced, or other data related to the worker. For example, worker data may be generated and / or updated e.g. periodically, iteratively, or upon employment of a worker. For example, operation 120 may include generating and / or maintaining a worker database. The worker data includes worker skill data indicative of worker skills, such as one or more skills of a worker, such as a worker skillset. For example, the worker skill data may be generated upon employment of the worker. In some embodiments, the worker may be interviewed and / or submit a resume and / or curriculum vitae indicative of the skills of the worker, and the worker data may be generated based on the interview results, resume and / or curriculum vitae of the worker.

[0030] According to some embodiments, the worker skill data of one or more workers may be generated based on a work history and / or a training history of the worker. For example, operation 120 may include receiving and / or obtaining data indicative of maintenance procedures and / or maintenance tasks the worker has performed, e.g. while being part of the workforce, and modifying the worker skill(s) associated with the maintenance procedures accordingly. For example, after having performed a supported maintenance procedure e.g. as defined in the dataset generated in operation 110, information indicating that the worker has performed the supported maintenance procedure may be recorded in operation 120. For example, the worker skill data may indicate that a worker has performed maintenance on a specific asset and / or performed a maintenance procedure for a defined number of times in a given timespan. For example, the worker skill data may indicate that the worker has received training on a specific asset and / or a specific maintenance procedure within a given timespan. Additionally, or alternatively, the worker skill data may be linked to the procedure dataset, e.g. to reflect that the worker has a (quantifiable) worker skill corresponding to a worker skill requirement, e.g. as defined in operation 112.

[0031] Likewise, operation 120 may include receiving and / or obtaining data indicative of training the worker has received, e.g. while being part of the workforce or before joining the workforce, and recording and / or modifying the worker skill(s) according to the maintenance procedures for which the worker has received training.

[0032] According to embodiments, the worker skill data may be indicative of one or more of an asset type proficiency of the worker, a maintenance task type proficiency of the worker, being e.g. indicative of the proficiency of one or more maintenance procedures, an environment type proficiency of the worker, being e.g. indicative of one or more environments, particularly harsh and / or dangerous environments in which the worker is experienced, and / or a cooperation proficiency of the worker, being indicative of e.g. a proficiency of the worker to operate in a team of workers. The asset type proficiency, the maintenance task type proficiency, the environment type proficiency and / or the cooperation proficiency may be indicative of a worker skill of the worker to perform maintenance procedures requiring the same or similar types of tasks and / or procedures. Accordingly, a worker skill, such as a competence score representing the worker skill, may be derived from one or more of the proficiencies for which the worker skill data may be indicative.

[0033] The method 100 includes, in operation 130, receiving a maintenance task indicative of a maintenance procedure to be performed. In some embodiments, the maintenance task may be indicative of several maintenance procedures to be performed, e.g. for maintaining an asset, or even several assets. The maintenance task may be a work order. The maintenance task may be a maintenance request, a failure report, a scheduled routine maintenance of one or more assets and / or industrial sites, or any other data construct indicative of one or more maintenance tasks.

[0034] According to some embodiments, the maintenance task may include a language-based description, such as a problem and / or task statement, indicative of one or more maintenance procedures. For example, the maintenance task may be text-based. In some embodiments, the maintenance procedure(s) indicated by the maintenance task may be determined based on an analysis determining a semantic similarity between the procedure data generated e.g. in operation 110, and the maintenance task or data included in the maintenance task. For example, operation 130 may include (semantically) matching terms, such as keywords, indicative of a maintenance procedure and / or an asset type to the procedure data, e.g. supported maintenance procedures included in the procedure data, to determine one or more maintenance procedures to be performed by a worker to fulfill the maintenance task.

[0035] For example, a maintenance task, which may be provided e.g. by a CMMS, may include text data such as “The motor of a pump assembly of type X is broken and needs to be replaced”. Semantic analysis may extract the asset type “pump assembly type X” and the maintenance procedure “motor needs to be replaced” from the text data so that the maintenance procedure may be determined, from the maintenance task, to include “replacing a motor of a pump assembly of type X”. Further analysis of the available procedure data may provide additional context, such as which type of motor is provided for the pump assembly of type X and / or which (further) maintenance procedures need to be performed when replacing the motor. The above example is given as an illustration, and should be understood as non-limiting. For example, according to some embodiments, additionally, or alternatively, supported maintenance procedures may be identifiable e.g. according to an identifier such as an index, and the maintenance task may directly reference one or more maintenance procedures to be performed e.g. by including the identifier.

[0036] According to embodiments, operation 130 may include determining one or more worker skills required for performing the maintenance procedure. In particular, the maintenance task may be indicative of one or more maintenance procedures. Each maintenance procedure may be indicative of one or more worker skill requirements. Accordingly, the required worker skills may be determined by accessing the procedure dataset, and particularly by retrieving, from the procedure dataset, the worker skill requirement(s) associated with the maintenance procedure(s) indicated by and / or determined from the maintenance task.

[0037] According to embodiments, matching a worker skill to the maintenance task may be performed according to a semantic similarity between a work history and / or a training history of the worker and the maintenance task. Likewise, maintenance procedures included in and / or indicated by the maintenance task may be matched to the worker dataset according to a semantic similarity between the maintenance task and a work history and / or a training history of the worker. Accordingly, based on the semantic analysis, the worker skill requirement, the corresponding worker skill, and the maintenance task may be linked according to one or more common maintenance procedures and / or identifiers indicative of a maintenance procedure.

[0038] According to embodiments, a worker skill may be matched by a contextual similarity to a maintenance procedure, particularly if an exact match is unavailable. For example, in a pool of workers, some workers may have a worker skill indicating a first maintenance procedure being similar to a second maintenance procedure, while no workers are available having a worker skill indicating the second maintenance procedure. The method may include determining a similarity between the first worker skill and the second worker skill. Accordingly, workers having the first worker skill may be matched for a maintenance procedure requiring the second worker skill. The method 100 includes, in operation 140, determining a competence score of the at least one worker. The competence score is determined based on the worker skill requirement of one or more maintenance procedures corresponding to the maintenance task, e.g. as determined in operation 130. For example, operation 130 may result in defining one or more worker skills required for performing the maintenance procedures corresponding to the maintenance task. Operation 140 further includes determining a worker skill, based on the worker skill data of the worker, e.g. as determined in operation 120. For example, the worker skill data may indicate a level of competence of a worker for the one or more maintenance procedures determined in operation 130.

[0039] According to embodiments, the maintenance task may be indicative of a plurality of maintenance procedures having different worker skill requirements. The competence score may be determined based on a plurality of worker skills corresponding to the plurality of different worker skill requirements. For example, operation 140 may determine a first competence score indicating that a worker has a high worker skill for a first maintenance procedure required for fulfilling the maintenance task, and a second competence score indicating that the worker has an average worker skill for a second maintenance procedure required for fulfilling the maintenance task. Accordingly, in some examples, the competence score may be a composite of the first competence score and the second competence score, such as e.g. a (weighted) average, and / or of one or more minimum competence scores of the worker. Additionally, or alternatively, a competence score may be determined individually for a plurality of workers and for a plurality of maintenance procedures. In some embodiments, e.g. if no one worker having sufficiently high competence scores for all maintenance procedures defined in the maintenance task can be selected, selecting the worker may include selecting a plurality of workers having different worker skills, and the plurality of workers may fulfil the maintenance task e.g. sequentially or as a team. For example, if multiple workers are available during the scheduled time of maintenance, they might be scheduled like a "combined worker" where one worker can compensate insufficient competences of the other worker(s). In a non-limiting example given here to help in understanding the invention, considering maintenance on a power train, one worker certified as electrician may inspect an LV switchgear, and another worker skilled in motor repair may check gearbox and motor.

[0040] The method 100 includes, in operation 150, selecting at least one worker based on the competence score. For example, according to a first embodiment, the worker may be selected based on a comparison of the competence scores of several workers so that a worker having the highest competence score among a pool of workers is selected. Beneficially, this may result in only skilled workers, or the most skilled worker among a pool of workers, performing one or more maintenance procedures to fulfill a maintenance task.

[0041] In operation 160, the selected worker performs the maintenance procedure, or several maintenance procedures, to fulfill the maintenance task. For example, following the selection in operation 140, the selected worker may be notified, travel to the asset, and perform the maintenance procedure(s) until the maintenance task is fulfilled.

[0042] While the above-described method 100 may beneficially result in skilled workers being selected for maintaining assets, the inventors of the present disclosure observe that, in some cases, the method may favor (over-)qualified workers, which may result in detrimental effects, such as an uneven distribution of workloads within the workforce, and / or the generation of entry barriers for less experienced workers. The following further embodiments may beneficially solve these problems at least in part.

[0043] According to embodiments, the method 100 may include determining, in operation 132, a task importance. Determining the task importance may include retrieving asset knowledge, such as information included in the procedure dataset, or receiving further information, such as assetspecific information, e.g. together with the maintenance task, or in addition to the maintenance task. Determining the task importance may include determining a criticality score.

[0044] According to embodiments, the criticality score may be indicative of a cost of the failure of the maintenance task. In the context of this disclosure, cost may be understood as a monetary cost, such as an expected loss in case of an unsuccessful maintenance. For example, in case of maintenance due to an asset having failed and / or being inoperable, the cost may be representative of the cost per time of the asset not being functional. For example, in case of routine maintenance, the cost may be indicative of a cost related to the unsuccessful maintenance itself, such as the time required by the worker, a cost of the work of the worker, such as an hourly or day rate, and / or travel costs. Likewise, cost may be understood as a nonmonetary cost, such as expectable disruptions of a workflow due to e.g. downstream effects of the asset not being maintained.

[0045] According to embodiments, the method 100, particularly generating the competence score in operation 140, may include determining a probability of a failure, or conversely the success, of a maintenance procedure being carried out by the worker based on a skill level of the worker, and the corresponding worker skill requirement. For example, data derived from the procedure dataset and data derived from the worker dataset may be matched and / or compared to determine if a worker from a pool of workers has a worker skill sufficiently high to successfully complete a maintenance procedure. In particular, determining a probability of a failure (or success) of a maintenance task for each worker may result in a metric, such as a score, indicative of whether the worker is expected to successfully complete a maintenance procedure. For example, for a highly skilled and / or experienced worker, generating the competence score may result in a competence score indicative of a 0 % or near 0 % probability of failure, while for an inexperienced worker, generating the competence score may result in a competence score indicative of non-zero, e.g. 50 % probability of failure. Beneficially, determining a probability of failure may facilitate selecting workers based on a competence score and / or competence gap of the worker.

[0046] According to embodiments, the method 100, particularly the generating the competence score in operation 140, may include determining a competence gap indicative of a skill level difference between the worker skill and the corresponding worker skill requirement. The worker skill requirement may be a metric indicating an expected worker skill for successfully completing the maintenance procedure. In some embodiments, the worker skill requirement may be a metric indicating that a worker having a determined worker skill level has a probability of task failure below a predetermined threshold. In some embodiments, the worker skill requirement may be derived from the procedure dataset. In some embodiments, the worker skill requirement may further be derived from, and / or be modified according to further metrics. In particular, the worker skill requirement may be modified according to a task importance, e.g. as determined in operation 132. For example, for a task having a low task importance, a larger competence gap may be permissible, i.e. a worker with a lower worker skill may be selected or selectable to perform the one or more maintenance procedures. Accordingly, the worker skill requirement may be modified, e.g. lowered. Likewise, for a task having a high task importance, the worker skill requirement may be increased accordingly.

[0047] According to embodiments, the method 100, particularly selecting the worker in operation 150, may include defining a limit indicative of a maximum allowable competence gap, and selecting a worker, e.g. from a set of available workers for which a competence score and / or competence gap was determined e.g. in operation 140, so that the worker has a competence score resulting in a competence gap not exceeding the limit.

[0048] Referring now to Fig. 2, further aspects of determining a worker competence and selecting a worker according to the worker competence will be explained. Fig. 2 shows a graph 200 in which the vertical axis represents worker experience of a worker performing a maintenance procedure, and the horizontal axis represents a worker competence. In the graph 200, a worker experience, which may be derived from the worker skill data, is represented as a time t the worker has spent performing the maintenance procedure and / or has spent training for performing the maintenance procedure. Accordingly, the axis t may represent, and / or be derived from, a work history and / or a training history of the worker. It should be noted that the graph 200 is simplified, and the axis t may not be a linear representation of time. For example, a time spent by a worker during specialized training may improve a worker competence more or less than the same time spent performing a maintenance procedure, and the difference is not reflected in the graph. According to some embodiments, instead of representing time, the axis t may represent another metric indicative of a worker experience, work history and / or training history of the worker, such as a number of completed maintenance procedures and / or tasks, or other metrics derivable from the worker skill data.

[0049] It should be noted that a worker having near zero experience at performing a maintenance procedure typically acquires at least some experience by training, and, once a sufficiently high worker competence is obtained through training, may gain further experience by performing the maintenance procedure on an asset, e.g. “in the field”. Furthermore, even an experienced worker may receive training, e.g. by attending training courses, apprenticing or the like, even when the worker’s competence is high. Accordingly, the axis t may represent the worker experience obtained by both training and performing the maintenance procedure.

[0050] The graph 200 represents a worker competence c for the maintenance procedure, which may be represented as a competence score. For example, the competence at the point may indicate a competence score of 50%, and the competence at the point c2may indicate a competence score of 100%. The competence score may be represented as an arbitrary number. For example, in some embodiments, the competence score may indicate a probability of failure and / or a probability of success when performing the maintenance procedure by the worker. For example, the competence score at the point may indicate an (estimated) probability of success of 50%, and the competence score at the point c2may indicate an (estimated) probability of success of 100%.

[0051] It should be noted that, in the example shown in Fig. 2, a worker having spent only limited time, i.e. no time, or little time t^ has a competence score of 0%. Accordingly, the probability of successfully completing the maintenance procedure by the worker is considered low or nearimpossible.

[0052] A worker having spent a time t2being longer than the time has a competence score of c^ such as e.g. 50%. Accordingly, the competence score may indicate that the worker has a 50% probability of successfully completing the maintenance procedure. A worker having spent a time t3being longer than the time t2has a competence score higher than c-i , such as e.g. 75%. Accordingly, in the example, the competence score higher than may indicate that the worker has a 75% probability of successfully completing the maintenance procedure. Accordingly, the worker may be described as more experienced than a worker having spent a lower time t2or tr

[0053] A worker having spent a time t4 being longer than the time t3has a competence score c2, such as e.g. 100%. Accordingly, the competence score c2may indicate that the worker has a 100% probability of successfully completing the maintenance procedure. Accordingly, the worker may be described as more experienced than a worker having spent a lower time t3it2or t^ The worker having spent the time t4may be considered an expert for performing the maintenance procedure. The competence score c2may be considered an expert competence score.

[0054] A worker having spent a time t5being longer than the time t4maintains a competence score c2, such as e.g. 100%. The graph 200 illustrates that, once a worker has an expert competence, further time spent on performing the maintenance procedure and / or training does not increase the competence of the worker.

[0055] It should be noted that the properties of the graph 200 may depend on the maintenance procedure, and may particularly be specific for several or each maintenance procedure. For example, less complex maintenance procedures may require lower times t for obtaining a high competence score, and even allow a non-zero competence score for low, near-zero or zero worker experience, while for more complex maintenance procedures, the worker competence c may be considered low or even zero, even though the worker has spent a high amount of time training for the maintenance procedure.

[0056] According to embodiments, selecting the worker, e.g. in operation 150 described with reference to Fig. 1 , may include defining a limit indicative of a maximum allowable competence gap. For example, in the graph 200 the limit L indicates a maximum allowable competence gap between the expert competence score c2and the actual competence score of the worker. For example, the limit L may indicate that, for the particular maintenance task, a competence score of e.g. 80% for the maintenance procedure defined by the maintenance task is sufficient. In some embodiments, the limit L may be indicative of the worker skill requirement and / or represent the worker skill requirement. According to embodiments, workers may be selected so that a selected worker has a competence score resulting in a competence gap not exceeding the limit. In the example shown in Fig. 2, from a pool of workers having varying competence scores, workers having a competence score of at least 80% may be chosen for the maintenance task, whereas for workers having a competence score of less than 80%, the competence gap exceeds the limit L.

[0057] According to embodiments, selecting the worker, e.g. in operation 150 described with reference to Fig. 1 , may include determining that the worker has a competence score higher than a prohibitive and / or minimal competence score and lower than a maximum competence score, such as the expert competence score c2, or a competence score lower than the expert competence score c2and higher than the prohibitive competence score. For example, as shown in Fig. 2, the prohibitive competence score may be defined by a limit, such as the limit L. Accordingly, in the embodiment, to be selected, the worker must have a competence score higher than the limit L. Furthermore, the worker must have a competence score lower than the maximum competence score, e.g. the expert competence score c2.

[0058] Beneficially, selecting workers having a competence score higher than a prohibitive competence score and lower than a maximum competence score may result in selecting workers for whom performing the maintenance procedure and / or the maintenance task can provide a worker skill improvement. In particular, performing the maintenance procedure may have a training effect. For example, when selecting a worker having spent e.g. a time t3for the maintenance procedure, successfully performing the maintenance procedure may increase the time spent by the worker, and consequently the worker may have a worker skill higher than at the time t3, and may further have a higher worker competence c as a result of having been selected and having performed the maintenance procedure.

[0059] According to embodiments, which may be combined with other embodiments described herein, selecting the worker, e.g. in operation 150 described with reference to Fig. 1 , may include determining a training effect score indicative of an improvement in worker skill for the selected worker when performing the maintenance procedure. In particular, the training effect score may be indicative of an expected improvement in worker experience and / or an expected improvement in worker competence. For example, for a highly skilled worker having a worker competence at or near the expert competence score c2, a training effect may be lower than for a worker being less experienced, such as a worker being near the prohibitive competence score. Accordingly, selecting the worker in operation 150 may include selecting a worker from a group of qualified workers for whom a higher training effect, or even the highest training effect, is obtainable when performing the maintenance procedure. Beneficially, determining the training effect score and selecting the worker according to the training effect score may result in selecting workers so that a worker skill and / or worker competence of the workforce is maximized. For example, determining the training effect score and selecting the worker according the training effect score may result in less workers who are considered experts at performing a maintenance procedure being selected for the maintenance procedure, and instead being selected for other maintenance procedures, which may beneficially broaden the skillset of the worker and / or the workforce.

[0060] According to embodiments, which may be combined with other embodiments described herein, selecting the worker, e.g. in operation 150 described with reference to Fig. 1 , may include applying a cost function, and selecting the worker, e.g. from a group of qualified workers, according to the cost function. For example, the cost function may determine expected costs in relation to a worker competence. For example, a worker having a competence score near the prohibitive competence score indicated by the limit L may be expected to require more time for completing the maintenance task than a worker having an expert competence score c2. Accordingly, the cost function may determine the cost of the prolonged maintenance procedure of the less competent worker. In some embodiments, the cost function may determine the cost as described herein with reference to the determining of the criticality score. In some embodiments, a metric provided by the cost function may be weighed against the training effect score. For example, the cost function may determine that, for a specific maintenance task, having the maintenance procedure(s) performed by a less experienced worker so that the less experienced worker may profit from the training effect of performing the maintenance procedure(s) outweigh the cost of a prolonged maintenance, and / or even an increase in a probability of task failure.

[0061] According to embodiments, the method 100 may include selecting, e.g. in operation 150 described with reference to Fig. 1 , a worker having a competence score lower than the limit L. Accordingly, the selected worker’s competence may be considered too low for performing the maintenance procedure, particularly lower than a prohibitive competence score. In the embodiment, the selected worker has a competence score higher than a minimum compensable competence score, indicated in Fig. 2 by a compensable limit CL, but lower than the limit L. For example, the limit L may define a minimum competence score of 80%, while the compensable limit CL may define a compensable minimum competence score of 75%. In the embodiment, before the selected worker performs the maintenance procedure, e.g. according to operation 160 described with reference to Fig. 1 , the method 100 includes training the selected worker to improve a worker skill of the selected worker required for the maintenance procedure. In particular, the method may include training the worker so that a worker skill, particularly a worker competence score, increases due to the training until the worker’s competence is above the minimum competence score defined by the limit L. Accordingly, in some embodiments, the competence gap of the selected worker may be larger than a competence gap in-between the expert competence c2and the prohibitive competence score defined by the limit L. In particular, a competence gap of the selected worker may be inbetween the compensable limit CL and the expert competence c2. According to some embodiments, determining a competence gap may include determining a difference between the worker’s competence and the prohibitive competence score, e.g. as indicated by the limit L in Fig. 2.

[0062] According to embodiments, the method 100 may include, for a selected worker having a competence score below a prohibitive competence score, e.g. as indicated by the limit L in Fig. 2, and above a minimum compensable competence score, e.g. as indicated by the compensable limit CL in Fig. 2, generating one or more training exercises based on the worker skill and the worker skill requirement. For example, worker skill data may indicate that the worker has a deficiency, such as a lack of experience or specific knowledge of a procedure, an asset, a tool and / or an environment, which may be compensated by generating training exercises and providing the training exercises to the worker. In some embodiments, the training exercises may include providing guides, manuals, or other types of study material to the worker. For example, training the worker may include selecting existing documentation or training modules annotated by (semantic) references to the procedure database and / or the worker database. In some embodiments, the training exercises may include video material, such as a training video. In some embodiments, the training exercises may include a simulation of the maintenance procedure, e.g. to be performed by the worker in a simulated environment. In some embodiments, the training exercises may include hands-on experience, e.g. having the worker attend and / or assist an experienced or even expert worker in the same or similar maintenance procedure. In some embodiments, training the worker may include providing a training plan generated for the worker in question, e.g. along with a deadline for the training. In some embodiments, the deadline might indicate an “instant training” to upskill for a present and urgent service need, or, alternatively, preparing for maintenance campaigns several months in the future.

[0063] According to embodiments, the method 100 may include defining a minimum compensable competence score, e.g. as indicated by the compensable limit CL shown in Fig. 2. For example, the minimum compensable competence score may be based on the prohibitive competence score, and be lowered according to an expected training effect of training the worker before performing the maintenance procedure. The expected training effect may be determined based on the available training exercises. For example, in case the training exercise is determined and / or expected to be highly effective, the difference between the minimum compensable competence score and the prohibitive competence score may be larger than when only limited training exercises are available.

[0064] According to embodiments, the method 100 may include determining a criticality score, e.g. according to embodiments described herein, and further include selecting the worker, e.g. in operation 150 described with reference to Fig. 1 , based at least in part on the criticality score. In particular, the method may include determining and / or adjusting the worker skill requirement, the prohibitive competence score and / or the maximum allowable competence gap according to the criticality score. For example, for a maintenance task having a high criticality score, e.g. for a maintenance task that is considered important and / or for which the cost of failure is high, the limit L may be shifted towards a higher competence score so that the worker selection becomes more selective. Accordingly, the compensable limit CL may be shifted towards a higher competence score, or even be set to be identical to the limit L. Likewise, for a maintenance task being considered less critical and / or important and / or having a lower criticality score, the limit L and / or the compensable limit CL may be shifted to be less restrictive to allow also less experienced workers.

[0065] Beneficially, when considering a training effect of the maintenance procedure, and / or an upskilling of workers having a competence score below a prohibitive limit and / or threshold, a more efficient use of an existing workforce may be obtained, thus reducing the cost and complexity of performing maintenance.

[0066] Referring now to Fig. 3, a data processing apparatus 300 is described. The data processing apparatus includes a database 310 with worker data of at least one worker, the worker data comprising worker skill data indicative of at least one of a plurality of worker skills. The data processing apparatus 300 includes a procedure database 320 with procedure data for supported maintenance procedures, the procedure data being indicative of at least one worker skill requirement defining a worker skill required for performing a respective maintenance procedure from the supported maintenance procedures. The data processing apparatus 300 includes an interface 340 configured for receiving a maintenance task indicative of a maintenance procedure to be performed. The data processing apparatus 300 includes a processor 330 configured for determining a competence score of the at least one worker, the competence score being determined based on the worker skill requirement of the maintenance procedure corresponding to the maintenance task, and the worker skill data of the worker. The data processing apparatus 300 is configured for selecting at least one worker based on the competence score; and generating an output indicative of the selected worker. According to embodiments, the data processing apparatus 300 may be configured for carrying out a method of selecting a worker, e.g. according to the method 100 described with reference to Fig. 1. For example, the data processing apparatus 300 may include a memory storing instructions, such as a computer program, that, when executed by the processor 330, cause the processor to carry out the method 100 described with reference to Fig. 1 , particularly the operations 110, 112, 120, 130, 132, 140 and 150, optionally with the exception of operation 160.

[0067] The data processing apparatus 300 may be implemented in the form of a computer and / or server. In particular, the data processing apparatus may be implemented in the form of a CMMS, or be implemented to communicate with a CMMS, e.g. for exchanging data between the CMMS and the data processing apparatus 300, e.g. via the interface 340.

[0068] Next, aspects of the present invention will be described by way of examples, which should not be understood as limiting.

[0069] The following pseudocode represents the calculations which may be carried out, according to embodiments, to determine a matching score M(...) and / or a competence gap Gap(...) of a worker. In the example, a maintenance procedure is determined from the values of the variables “task” and “asset_type”.

[0070] H(worker, task, asset_type) is defined as the hours of experience of worker, and may correspond to the worker experience, e . g . t, described with reference to Fig . 2;

[0071] H_opt (task, asset_type) is defined as the hours expected for full success, and may correspond to the expert competence score, e . g . t at c2, described with reference to Fig . 2;

[0072] H_min (task, asset_type) is defined as the minimal hours acceptable to execute task, and may correspond to the limit, e . g . L, described with reference to Fig . 2;

[0073] M(worker, task, asset_type) = { 100% if H(...) >H_opt(...)

[0074] { 100%*H / H_opt if H(...)<H_opt(...)

[0075] Gap(worker, task, asset_type) = 100%-M(...) ; from perspective of H_opt Gap_max(...) = (H_opt(...) -H_min(...) ) / H_opt * 100%

[0076] The following pseudocode represents the calculations which may be carried out, according to embodiments, to determine if a competence gap of a worker having a competence score below a prohibitive competence score, such as the limit L described with reference to Fig. 2, can be overcome through training.

[0077] Training_effect(task, asset_type) [0...100%] ; compensating effect of training

[0078] M_projected = { 100% if M(...) + Training_effect(...) > 100%

[0079] {M(...) + Training_effect(...)

[0080] Only workers are chosen for which M_projected(...) > 100%-Gap_max(...)

[0081] Training is also possible for M(...) > 100%-Gap_max(...) while M(...)< 100%

[0082] The results of determining a matching score are shown as an example for the following two exemplary cases in which the maintenance procedure defines the repair of a motor. For a skilled worker, the following metrics are determined:

[0083] H(skilled_worker, repair, motor) = 80h

[0084] H_opt( repair, motor) = 60h

[0085] H_min( repair, motor) = 40h

[0086] M(skilled_worker, repair, motor) = 100%

[0087] Gap(...)=0%

[0088] M_projected(skilled_worker, repair, motor) = 100%

[0089] Upskill_opportunity(...)=0%

[0090] For a less skilled worker, the following metrics are determined:

[0091] H(less_skilled_worker, repair, motor) = 30h

[0092] H_opt( repair, motor) = 60h H_min ( repair , motor) = 40h

[0093] M(less_skilled_worker, repair, motor) = 50%

[0094] Gap(...)=50%

[0095] Gap_max(...) = 33%

[0096] Train ing_ef feet ( repair, motor) =20%

[0097] M_projected(less_skilled_worker, repair, motor) = 70%

[0098] Upskill_opportunity(...)=20%

[0099] Accordingly, for the skilled worker, a good match between the maintenance procedure (repair, motor) was found, as represented by the metric M. For the less skilled worker, the metric M was determined to be below a prohibitive threshold defined by the value Gap_max of 33%. It was, however, determined that training of the less skilled worker may increase the worker skill of the less skilled worker by 20%, resulting in a worker competence M_projected of 70% being within the maximum competence gap. Accordingly, the less skilled worker may be selected for fulfilling the maintenance task after training. The matching may be dependent on the cost function, taking into account both the upskill_opportunity of the less skilled worker, and the cost of training and potentially delayed maintenance.

[0100] The following pseudocode represents the calculations which may be carried out, according to embodiments, to determine a training effect of performing a maintenance procedure, such as determining the training effect score indicative of an improvement in worker skill for the selected worker according to embodiments described herein.

[0101] T_execution(worker, task) > 0; representing the task execution time

[0102] T_training_effect(task) <0; representing a compensation of used time due to training

[0103] T_training_aware_execution (worker, task) = T_execution(worker, task) + T_training_ effect(worker, task) ; updated metric to record in the worker dataset for the worker

[0104] In an example, a skilled worker may require 0.5 hours to perform the maintenance, while a less-skilled worker may require 1.5 hours. Provided that the T_training_effect metric is -1 or larger, the training effect for a less-skilled worker compensates the longer expected repair time. Accordingly, when selecting the worker while implementing a cost function, the less-skilled worker may be chosen due to the added training effect.

[0105] While the foregoing is directed to certain embodiments of the present disclosure, other and further embodiments may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

Claims

Claims:1 . Method of fulfilling a maintenance task by a worker, comprising:- receiving a maintenance task indicative of a maintenance procedure to be performed;- generating a worker dataset comprising worker data of at least one worker, the worker data comprising worker skill data indicative of at least one of a plurality of worker skills;- generating a procedure dataset comprising procedure data for maintenance procedures, the procedure data being indicative of at least one worker skill requirement defining a worker skill required for performing a respective maintenance procedure from the supported maintenance procedures;- determining a competence score of the at least one worker, the competence score being determined based on the worker skill requirement of the maintenance procedure corresponding to the maintenance task, and the worker skill data of the worker;- selecting at least one worker based on the competence score;- performing the maintenance procedure by the selected worker.

2. The method according to claim 1 , wherein the worker skill data of the worker is generated based on a work history and / or a training history of the worker.

3. The method according to claim 1 or 2, wherein generating the competence score comprises determining a competence gap indicative of a skill level difference between at least one worker skill and at least one corresponding worker skill requirement.

4. The method according to any one of the preceding claims, wherein generating the competence score comprises determining a probability of a failure, or conversely the success, of the maintenance procedure based on a skill level difference of the worker skill and the corresponding worker skill requirement.

5. The method according to any one of claims 3 or 4, wherein selecting the worker comprises defining a limit indicative of a maximum allowable competence gap; andselecting a worker having a competence score resulting in a competence gap not exceeding the limit.

6. The method according to any one of the preceding claims, wherein selecting the worker comprises determining that the worker has a competence score higher than a prohibitive competence score and lower than a maximum competence score.

7. The method according to claim 6, further comprising determining a training effect score indicative of an improvement in worker skill for the selected worker when performing the maintenance procedure.

8. The method according to any one of the preceding claims, further comprising determining that the selected worker has a competence score higher than a minimum compensable competence score, wherein the minimum compensable competence score is indicative of a prohibitive competence score being compensable by training; and before performing the maintenance by the selected worker, training the selected worker to improve a worker skill of the selected worker required for the maintenance procedure.

9. The method according to claim 8, wherein training the worker comprises generating one or more training exercises based on the worker skill and the worker skill requirement.

10. The method according to any one of the preceding claims, wherein selecting the worker comprises defining a criticality score, the criticality score being indicative of a cost of the failure of the maintenance task, and wherein the worker skill requirement, prohibitive competence score and / or or maximum allowable competence gap is adjusted according to the criticality score.

11. The method according to any one of the preceding claims, wherein the maintenance task is indicative of a plurality of maintenance procedures having different worker skill requirements, and wherein the competence score is determined based on a plurality of worker skills corresponding to the plurality of different worker skill requirements.

12. The method according to any one of the preceding claims, wherein the worker skill data is indicative of at least one selected from the group consisting of:- an asset type proficiency;- a maintenance task type proficiency;- an environment type proficiency;- a cooperation proficiency.

13. The method according to any one of the preceding claims, further comprising: matching at least one maintenance procedure to the maintenance task according to a semantic similarity between the procedure data and the maintenance task.

14. The method according to any one of the preceding claims, further comprising: matching a worker skill to the maintenance task according to a semantic similarity between a work history and / or a training history of the worker and the maintenance task.

15. A data processing apparatus comprising: a database comprising worker data of at least one worker, the worker data comprising worker skill data indicative of at least one of a plurality of worker skills; a procedure database comprising procedure data for supported maintenance procedures, the procedure data being indicative of at least one worker skill requirement defining a worker skill required for performing a respective maintenance procedure from the supported maintenance procedures; an interface configured for receiving a maintenance task indicative of a maintenance procedure to be performed; a processor configured for determining a competence score of the at least one worker, the competence score being determined based on the worker skill requirement of the maintenance procedure corresponding to the maintenance task, and the worker skill data of the worker; whereinthe data processing apparatus is configured for selecting at least one worker based on the competence score; and generating an output indicative of the selected worker.

Images