Condition evaluation of an elevator

Abstract

A method for evaluating a condition of an elevator (100) is provided, the method comprises: receiving (210) a number of indicators indicative of at least one state occurring in the elevator (100); generating (220) a fingerprint by aggregating the number of indicators, the fingerprint being descriptive of the condition of the elevator (100) at an instant of time; determining (230) one or more matches in a predefined accuracy between the fingerprint descriptive of the condition of the elevator (100) generated at the instant of time and at least one other fingerprint descriptive of a condition of an elevator (100) at at least one other instant of time; and generating (240) an output comprising data descriptive of a prediction of a future condition of the elevator (100). A computing apparatus (120) and a computer program are provided.

Description

[0001] CONDITION EVALUATION OF AN ELEVATOR

[0002] TECHNICAL FIELD

[0003] The invention concerns in general the technical field of elevators. More particularly, the invention concerns an elevator maintenance.

[0004] BACKGROUND

[0005] There are developed various approaches for monitoring and diagnosing an operation of an elevator. One approach is to use so-called state values, such as fault codes, which are generated to indicate an operating state of an entity in the elevator. The state value may also carry data defining in more detail a type of the state the respective entity is experiencing. It may be arranged that a computing entity, such as an elevator controller, is configured to receive the state values from a number of sources, i.e. from sub-entities of the elevator system, and the state values are analyzed and necessary actions are initiated based on the result of the analysis. The actions may e.g. comprise a prevention of a use of the elevator and / or generating a maintenance call for the elevator e.g. in a situation that the elevator is experiencing a serious fault state.

[0006] The existing solutions are operative to some extent but there is a need to introduce further sophisticated solutions that increase understanding of a condition of an elevator and a development of the condition.

[0007] SUMMARY

[0008] The following presents a simplified summary in order to provide basic understanding of some aspects of various invention embodiments. The summary is not an extensive overview of the invention. It is neither intended to identify key or critical elements of the invention nor to delineate the scope of the invention. The following summary merely presents some concepts of the invention in a simplified form as a prelude to a more detailed description of exemplifying embodiments of the invention. An object of the invention is to present a method, a computing apparatus and a computer program for evaluating a future condition of an elevator.

[0009] The objects of the invention are reached by a method, a computing apparatus and a computer program as defined by the respective independent claims.

[0010] According to a first aspect, a method for evaluating a condition of an elevator is provided, the method, performed by a computing apparatus, comprises: receiving a number of indicators indicative of at least one state occurring in the elevator, generating a fingerprint by aggregating the number of indicators, the fingerprint being descriptive of the condition of the elevator at an instant of time, determining one or more matches in a predefined accuracy between the fingerprint descriptive of the condition of the elevator generated at the instant of time and at least one other fingerprint descriptive on a condition of an elevator at at least one other instant of time, generating, in response to determining the one or more matches in the predefined accuracy, an output comprising data descriptive of a prediction of a future condition of the elevator based on history data descriptive of a development of a condition of the elevator whose fingerprint generated the match with the generated fingerprint.

[0011] The number of indicators forming the fingerprint may be a predefined set of state values obtainable from the elevator.

[0012] For example, the fingerprint may be a heatmap representation.

[0013] The determination of the one or more matches may be performed with a machine-learning model trained for the task. For example, the machinelearning model may be a transformer model. Furthermore, the aggregation of the number of indicators may be performed with an embedding operation. Still further, the generation of the output comprising data descriptive of the prediction of the future condition of the elevator may be performed based on at least one of the following: at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the same elevator to that fingerprint that generated the one or more matches with the generated fingerprint; metadata associated with at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the same elevator to that fingerprint that generated the one or more matches with the generated fingerprint wherein the metadata being descriptive of a condition of the elevator at an instant the respective fingerprint is generated.

[0014] According to a second aspect, a computing apparatus for evaluating a condition of an elevator is provided, the computing apparatus is configured to perform: receive a number of indicators indicative of at least one state occurring in the elevator, generate a fingerprint by aggregating the number of indicators, the fingerprint being descriptive of the condition of the elevator at an instant of time, determine one or more matches in a predefined accuracy between the fingerprint descriptive of the condition of the elevator generated at the instant of time and at least one other fingerprint descriptive on a condition of an elevator at at least one other instant of time, generate, in response to determining the one or more matches in the predefined accuracy, an output comprising data descriptive of a prediction of a future condition of the elevator based on history data descriptive of a development of a condition of the elevator whose fingerprint generated the match with the generated fingerprint. The computing apparatus may be configured to apply a predefined set of state values obtainable from the elevator as the number of indicators forming the fingerprint.

[0015] For example, the computing apparatus may be configured to generate a heatmap representation as the fingerprint.

[0016] Further, the computing apparatus may be configured to perform the determination of the one or more matches with a machine-learning model trained for the task. For example, the machine-learning model may be a transformer model executed by the computing apparatus. The computing apparatus may be configured to perform the aggregation of the number of indicators with an embedding operation.

[0017] Still further, the computing apparatus may be configured to perform the generation of the output comprising data descriptive of the prediction of the future condition of the elevator based on at least one of the following: at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the same elevator to that fingerprint that generated the one or more matches with the generated fingerprint; metadata associated with at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the same elevator to that fingerprint that generated the one or more matches with the generated fingerprint wherein the metadata being descriptive of a condition of the elevator at an instant the respective fingerprint is generated.

[0018] According to a third aspect, a computer program is provided, the computer program comprising instructions to cause the computing apparatus according to the second aspect as defined above to execute the steps of the method according to the first aspect as defined above.

[0019] The expression "a number of” refers herein to any positive integer starting from one, e.g. to one, two, or three. The expression "a plurality of” refers herein to any positive integer starting from two, e.g. to two, three, or four.

[0020] Various exemplifying and non-limiting embodiments of the invention both as to constructions and to methods of operation, together with additional objects and advantages thereof, will be best understood from the following description of specific exemplifying and non-limiting embodiments when read in connection with the accompanying drawings.

[0021] The verbs “to comprise” and “to include” are used in this document as open limitations that neither exclude nor require the existence of unrecited features. The features recited in dependent claims are mutually freely combinable unless otherwise explicitly stated. Furthermore, it is to be understood that the use of “a” or “an”, i.e. a singular form, throughout this document does not exclude a plurality.

[0022] BRIEF DESCRIPTION OF FIGURES

[0023] The embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings.

[0024] Figure 1 illustrates schematically an implementation for evaluating a condition on an elevator according to an example.

[0025] Figure 2 illustrates schematically a method according to an example.

[0026] Figure 3 illustrates schematically examples of an aggregation of a number of indicators.

[0027] Figure 4 illustrates schematically an apparatus according to an example.

[0028] DESCRIPTION OF THE EXEMPLIFYING EMBODIMENTS

[0029] The specific examples provided in the description given below should not be construed as limiting the scope and / or the applicability of the appended claims. Lists and groups of examples provided in the description given below are not exhaustive unless otherwise explicitly stated. Figure 1 illustrates schematically an implementation, or a system, for evaluating a condition of an elevator 100 according to an embodiment of the invention. The elevator 100 may refer to a single elevator or to a system consisting of a plurality of elevators thus forming an elevator group. The elevator 100 may comprise a number of sub-systems 110A, 110B, 110C that may relate to an operation of the elevator. As non-limiting examples of the subsystems 110A, 110B, 110C an elevator door and an elevator braking system may be mentioned to. The sub-systems 110A, 110B, 110C may comprise a plurality of entities wherein a controller controls at least in part the operation of the respective sub-system 110A, 110B, 110C. The controller may also be configured to gather and manage data obtained in the respective sub-system 110A, 110B, 110C. In accordance with at least some embodiments the data of the sub-system 110A, 110B, 110C may be an indicator that is indicative of at least one state occurring in the elevator. The data may be obtained directly in the sub-system 110A, 110B, 110C or it may be derived from data available in the sub-system 110A, 110B, 110C. For example, the state may be indicative of the state of the respective sub-system 110A, 110B, 110C. In some embodiments of the invention the indicator may define one or more fault codes of the respective sub-system 110A, 110B, 110C as the indicator in a situation that the sub-system 110A, 110B, 110C is experiencing an error state.

[0030] The number of sub-systems 110A, 110B, 110C may be communicatively connected to a computing apparatus 120. The computing apparatus 120 is arranged to receive data from the number of sub-systems 110A, 110B, 110C over a communication channel established between the entities. The communication channel may be based on, and thus implemented with, wired or wireless communication technology. For sake of clarity it is worthwhile to mention that the communication channel may be established directly between each sub-system 110A, 110B, 110C and the computing apparatus 120 or indirectly so that there is a further entity between the at least one sub-system 110A, 110B, 110C and the computing apparatus 120 wherein the further entity in the described situation may e.g. be an elevator controller (not shown in Figure 1 ). It may also be arranged that one of the sub-systems 110A, 110B, 110C is arranged to gather data from at least one other sub-system 110A, 110B, 110C and to deliver that data to the computing apparatus 120. Thus, the computing apparatus 120 may receive one or more indicators, or the raw data, from the one or more sub-systems 110A, 110B, 110C. In case it receives raw data the computing apparatus 120 may be configured to generate one or more indicators indicative of the at least one state occurring in the elevator 100 on the basis of raw data. Thus, the computing apparatus 120 possesses, at some point of time, the number of indicators. For avoidance of any doubt it is hereby mentioned that it may also be arranged that the elevator controller is configured to operate in the role of the computing apparatus 120 and thus belongs to the elevator system 100 by default.

[0031] In order to describe the invention and its various embodiments further it is referred to Figure 2. Figure 2 schematically illustrates a method according to an example wherein the method may be executed by the computing apparatus 120 as described herein. In step 210, the computing apparatus 120 receives 210 a number of indicators indicative of at least one state occurring in the elevator as described in the foregoing description among other description herein. The number of indicators may represent the state of one or more entities of the elevator 100. For example, an indicator may represent a state of an elevator door system whereas another indicator may represent a state of an elevator braking system. In some embodiment of the invention the indicator is generated and provided to the computing apparatus 120 in every case, e.g. even if the respective entity, or entities, is operating as expected. In another embodiment, the indicator may only be generated in situations when the respective entity, or entities, is malfunctioning. It may be arranged that the computing apparatus 120 is arranged to receive the number of indicators from a number of sources, i.e. from the sub-system(s) 110A, 110B, 110C substantially at the same instant of time, or in a time window, in order to describe the situation occurring in the elevator 100 at the instant of time. In some other embodiment the computing apparatus 120 may receive the indicators over a predefined period of time. The receipt 210 of indicators may be initiated by generating a trigger signal by the computing apparatus 120 to the number of sub-systems 110A, 11 OB, 110C to cause the number of subsystems 110A, 110B, 110C to respond and to provide the number of indicators, i.e. the data representing the indicator value(s), to the computing apparatus 120.

[0032] In response to the receipt 210 of the number of indicators the computing apparatus 120 is configured to generate 220 a fingerprint by aggregating the number of indicators. Thus, the generated fingerprint is descriptive of the condition of the elevator 100 at an instant of time the fingerprint is generated. This is possible because the indicators represent the state, or states, of the number of sub-systems 110A, 110B, 110C and, thus, enable forming a fingerprint on the condition of the elevator 100. Figure 3 illustrates as a nonlimiting manner some examples of an aggregation of the number of indicators to generate 220 the fingerprint. Hence, the fingerprints may be seen as a vectorized representation based on the one or more indicators. The computing apparatus 120 may be configured to maintain data structure into which the number of indicators are stored in response to the receipt of the indicators, or generation of the indicators, at various instants, e.g. in time or instants of measurement (cf. columns denoted with Instant 1 , Instant 2, Instant 3, Instant 4). For example, the new inputs received by the computing apparatus 120 are stored to a further column in the data structure in a continuous manner. The indicators may be stored with respect to each sub-system (cf. rows denoted with Sub-system 1 , Sub-system 2, Sub-system 3, ... , Sub-system N). In the table illustrated in Figure 3 the indicators are denoted with x. For sake of clarity it is worthwhile to bring out that the indicators, i.e. the values of the indicators, may differ from each other in order to describe details with respect to the state of the respective sub-system even if they are denoted with x in Figure 3. As such, the indicator values may be predefined default values for the specific sub-system and selected in accordance with the state of the respective subsystem in the instant in question or their value may define the situation the respective sub-system is experiencing (cf. e.g. a measurement value). In some embodiments of the invention the indicator may be expressed with one or more default values even independently of the sub-system, such as the selectable values are defined so that a first value indicates that the sub-system in question operates properly and a second value indicates an error state. Any other approaches may be taken with respect to the indicator values. Moreover, the gathering of information in the described manner in the data structure enables the generation 220 of the fingerprint. In accordance with some embodiments the fingerprint may be a sub-portion of the data structure, e.g. consisting of the indicator values of one or more instants. Figure 3 shows two non-limiting examples of fingerprints derivable from the data structure wherein the fingerprints comprise an aggregation of the number of indicators. The derived fingerprints are referred with A and B in Figure 3. In some embodiments, the fingerprint may be an aggregation of indicators of some of the sub-systems, not all of them. In any case, the fingerprint provides information describing the condition of the elevator based on the information defined by the indicators indicative of the one or more states occurring in the elevator 100. In some embodiments, the generation of the fingerprint comprises a generation of a heatmap as the fingerprint wherein the heatmap is generated over a predefined length of time. Thus, the heatmap aggregates the number of indicators received by the computing apparatus 120 and visualizes a magnitude of individual values within a dataset as a color expressed e.g. by hue or intensity as non-limiting examples. This kind of fingerprint effectively brings out the active areas within the monitored environment.

[0033] In response to the generation 220 of the fingerprint the computing apparatus 120 is configured to perform a determination 230 of one or more matches in a predefined accuracy between the fingerprint descriptive of the condition of the elevator 100 as generated and at least one other fingerprint descriptive on a condition of an elevator at at least one other instant of time. The determination 230 may be implemented so that the computing apparatus 120 is provided with an access to data storage 130 configured at least to store fingerprints descriptive of conditions of a number of elevators at instants of time being earlier than condition of the elevator at the instant of time the fingerprint in question is generated 220. In other words, the data storage 130 may be arranged to store history data of one or more elevators, such as the history data of the elevators 100 in question as well as history data of other elevators than the one for which the fingerprint in question is generated 220. The history data is advantageously stored in a form of fingerprint applicable to be used in searching possible match(es) with the generated fingerprint. Furthermore, the stored history data may comprise so-called metadata together with the fingerprint wherein the metadata may comprise at least some data obtained, or received, from the elevator at the instant the fingerprint is generated. In other words, the metadata may be data that is descriptive of the condition of the elevator in question at the instant of time the fingerprint is generated to, and, possibly, the data based on which the fingerprint is generated to. For example, the metadata may be associated with the generated fingerprint in the data storage so that it is inquirable, i.e. possible to be inquired, therefrom with or without the inquiry with respect to the fingerprint. The metadata may e.g. comprise description on a condition, e.g. a type of one or more issues (e.g. component information, root cause, measurement data, etc.), in the elevator and / or a duration how long after the previous fingerprint the issue occurred. For avoidance of doubt it is worthwhile to mention that the data storage 130 may be an external entity to the computing apparatus 120 or an internal entity, such as a memory unit storing the reference data, i.e. the history data. The search of the one or more matches may e.g. be implemented by generating an inquiry to the data storage 130 wherein the data storage provides a response indicating if the one or more matches is found. The response indicating the match may also comprise further data that is available in the data storage 130 wherein the further data may be related to the elevator whose fingerprint generated the one or more matches with the fingerprint in question. The further data may e.g. comprise the at least some of the further fingerprints, i.e. the following fingerprints from the one generated the match, of the respective elevator generated after the fingerprint that generated the one or more matches and that are stored in the data storage 130. In other words, the one or more fingerprints received as the further data define a development in the condition of the elevator in question. For avoidance of doubt, the determination of the one or more matches 230 may be set to occur in a predefined accuracy. This refers to an approach that it may be required that the indication of the one or more matches is only generated when the determination exceed a predefined reference level. It may be required that the match is 100 %, but other reference level may also be set. For example, the reference level may be set to 80 % and if that is exceeded the indication of the match is generated. The determination operation may according to an embodiment comprise a comparison of features of the fingerprints in a predefined manner in order to generate a result of the determination 230. At this stage it may be mentioned that if no match is determined, the computing apparatus may continue its operation by receiving 210 further indicators and iterating the process as described.

[0034] In response to the determination 230 of the one or more matches in the predefined accuracy, an output may be generated 240 wherein the output comprises data descriptive of a prediction of a future condition of the elevator. The prediction may be generated based on history data descriptive of a development of a condition of the elevator whose fingerprint generated the match with the generated fingerprint. This is because the computing apparatus 120 has the access to further data stored with respect to the elevator that generated the one or more matches and, thus, the further data, such as later fingerprints, following the fingerprint that generated the match may be utilized in the generation of the prediction on the development of the condition. Alternatively or in addition, the further data refers to the metadata stored together with the stored fingerprints. Hence, the computing apparatus 120 may be configured to derive from data relating to the elevator whose fingerprint generated the one or more matches with the generated fingerprint the prediction on the development of the condition of the elevator 100 in question. Such data may e.g. define the development of the state(s) occurring in the elevator 100 on the basis of the fingerprint. For example, the computing apparatus 120 may decrypt the fingerprint, or fingerprints, with a reverse operation to an operation the fingerprint is generated and, thus, to determine the indicator values of the fingerprint(s). In some other approaches, the computing apparatus 120 may receive, from the data storage, the metadata relating to a number of fingerprints generated from the elevator(s) that generated the one or more matches with the fingerprint in question and analyse the metadata in order to generate the prediction on the development of the state occurring in the elevator. Thus, by assuming that the elevator under evaluation follows the development of the elevator whose fingerprint generated the one or more matches, the future condition of the elevator 100 under evaluation may be determined on the basis of the decrypted indicator value(s), and / or the associated metadata, and the output may be generated 240 accordingly. The generation 240 of the output may e.g. refer to that a signal is generated, e.g. in a form of a predefined message, in order to cause the information visible to a responsible party, such as to a maintenance department and to persons therein. The output, such as the prediction, may also comprise data defining a time window for executing the maintenance in order to keep the elevator 100 in operation as efficiently as possible. Alternatively or in addition, the output may define a maximum duration the elevator 100 may be kept in operation in a safe manner in view of the current condition and its development. For example, the prediction may be dependent on a duration of an applicability of one or more components detected to start misbehaving so as to provide an output how long the elevator in question may be used.

[0035] In some embodiments the fingerprints and any associated data, such as the metadata, stored in the data storage are only such that are detected to define an error state of the elevator. These states may e.g. have led to a service call or to a prevention to operate the elevator. For example, if one such state is detected in an elevator, the further fingerprints, and any associated data, from the same elevator may also advantageously be stored in order to enable the prediction of the development in some other elevators. It may also be arranged that if one or more matches is found between a fingerprint of an elevator and a fingerprint stored in the data storage, the situation may be followed by determining if the following fingerprint from the same elevator and the following fingerprint of the elevator that generated the one or more matches with the previous at least one fingerprint still match. In such a case the probability is increased that the development of the elevator in question follows the development of the reference elevator. The same may be iterated in response to a generation of a new fingerprint and if the match continues to be present, the prediction becomes more accurate.

[0036] In the foregoing description at least some aspects of the invention are substantially described in an implementation wherein the computing apparatus 120 is configured to execute a number of so-called rule-based algorithms to perform the method in the manner as described. In addition to such an approach, the invention may be implemented with an applicable machinelearning model trained for the task. In accordance with an embodiment of the invention a so-called transformer model may be trained to generate an output descriptive of a prediction of a future condition of the elevator. The training of the transformer model may be performed with a training data set that at least comprises fingerprints generated by one or more computing apparatuses 120 from data received by the respective computing apparatuses 120 over a time. Hence, the fingerprints used in the training data set may represent fingerprints of a number of elevators 100 and at least so that a plurality of fingerprints generated chronologically consecutively are from the same elevator 100 in order to use training data that is descriptive on the development of a condition of the elevator. Hence, having such data from a plurality of elevators 100 the trained machine-learning model may generate a detection to various situations. Moreover, the training data is advantageously also such that it is received from, i.e. represents, elevator(s) which correspond to the elevator(s) 100 under evaluation at a predefined accuracy. This improves the accuracy of the evaluation in the described manner. The training of the machine-learning model being e.g. the transformer model may be implemented with known training mechanisms, such as by applying so-called supervised or unsupervised learning approach. The description relating to the training of the machine-learning model also covers an implementation wherein the training is performed at so-called lower dimensional representation of the training data, such as by using counts of indicators over a predefined period of time, such as over a day. Thus, such embedded data may be used as the fingerprint in the training. Correspondingly, the fingerprint generated by aggregating the number of indicators to represent the condition of the elevator at the instant of time is naturally generated with the corresponding embedding mechanism as the training data.

[0037] The machine-learning model trained with the embedding approach as described is beneficial in a sense that it enables a prediction of the future condition of the elevator, and, thus, the future indicators, even if the current state represented by the current fingerprint does not generate any match. For example, it is possible to generate fingerprints through the embedding over a period of time, e.g. over a week, that then to make the machine-learning model to determine a pattern from the past (cf. the training with history data) that is most similar to the fingerprint in question and, then, use that pattern to predict the development. Thus, it may be considered that the pattern and the aggregated indicators by embedding them over the period of time form comparable fingerprints as such that enable the prediction if the one or more matches is found in a predefined accuracy. The machine-learning model may also be configured to use the metadata in the generation of the output in the manner as described in the foregoing description upon the determination of one or more matches as described. The machine-learning model may also be trained to generate more sophisticated output comprising data descriptive of the prediction of the future condition of the elevator wherein the output may e.g. be derived by applying one or more so-called large language models (LLM), i.e. the output is enriched with the models as mentioned.

[0038] An example of an apparatus configurable to implement the operation of the computing apparatus 120 is schematically illustrated in Figure 4. The apparatus may be configured to perform the method according to the invention as described with the examples in the foregoing description. For sake of clarity, it is worthwhile to mention that the block diagram of Figure 4 depicts some components of an apparatus that may be employed to implement a functionality of the computing apparatus 120 as described. The apparatus of Figure 4 comprises a processing unit 410 comprising one or more processors and a memory unit 420. The memory unit 420 may store data, such as the received indicators and any other data, also computer program code 425 causing the operation in the described manner. The computer program code 425 may also comprise a trained machine-learning model. In at least some embodiments, the apparatus may further comprise a communication interface 430, such as a wireless communication interface or a communication interface for wired communication, or both, to communicate with other entities as described. The communication interface 430 may thus comprise one or more modems, antennas, and any other hardware and software for enabling an execution of the communication e.g. under control of the processing unit 410. Furthermore, I / O (input / output) components may be arranged, together with the processing unit 410 and a portion of the computer program code 425, to provide a user interface for receiving input from a user, such as from a technician, and / or providing output to the user of the apparatus when necessary. In particular, the I / O components may include user input means, such as one or more keys or buttons, a keyboard, a touchscreen, or a touchpad, etc. The I / O components may include output means, such as a loudspeaker, a display, or a touchscreen. The components of the apparatus may be communicatively connected to each other via data bus that enables transfer of data and control information between the components.

[0039] The memory unit 420 and at least a portion of the computer program code 425 stored therein may further be arranged, with the processing unit 410, to cause the apparatus to perform at least a portion of a method as is described herein. The processing unit 410 may be configured to read from and write to the memory unit 420. Although the processing unit 410 is depicted as a respective single component, it may be implemented as respective one or more separate processing entities. Similarly, although the memory unit 420 is depicted as a respective single component, it may be implemented as respective one or more separate components, some, or all of which may be integrated I removable and I or may provide permanent I semi-permanent I dynamic I cached storage. The computer program code 425 may comprise computer-executable instructions that implement functions that correspond to steps implemented in the method when loaded into the processing unit 410 of the respective control system. As an example, the computer program code 425 may include a computer program consisting of one or more sequences of one or more instructions. The processing unit 410 is able to load and execute the computer program by reading the one or more sequences of one or more instructions included therein from the memory unit 420. The one or more sequences of one or more instructions may be configured to, when executed by the processing unit 410, cause the apparatus, such as a computer, to perform a method as described. Hence, the apparatus may comprise at least one processing unit 410 and at least one memory unit 420 including the computer program code 425 for one or more programs, the at least one memory unit 420 and the computer program code 425 configured to, with the at least one processing unit 410, cause the apparatus implementing the computing apparatus 120 to perform the method.

[0040] The computer program code 425, or at least some portion of it, may be provided e.g. a computer program product comprising at least one computer- readable non-transitory medium having the computer program code 425 stored thereon, which computer program code 425, when executed by the processing unit 410 causes the computing apparatus 120 to perform the method. The computer-readable non-transitory medium may comprise a memory device or a record medium, such as a CD-ROM, a DVD, a Blu-ray disc, or another article of manufacture that tangibly embodies the computer program. As another example, the computer program may be provided as a signal configured to reliably transfer the computer program.

[0041] Still further, the computer program code 425 may comprise a proprietary application, such as computer program code for causing an execution of the method in the manner as described in the description herein. Any of the programmed functions mentioned may also be performed in firmware or hardware adapted to or programmed to perform the necessary tasks.

[0042] For sake of completeness it is worthwhile to mention that the entity performing the method in the role of the computing apparatus 120 may also be implemented with a plurality of apparatuses, such as the one schematically illustrated in Figure 4, as a distributed computing environment. For example, one of the apparatuses may be communicatively connected with the other apparatuses, and e.g. share the data of the method, to cause another apparatus to perform at least one other portion of the method. As a result, the method performed in the distributed computing environment generates the control interface as described. The functionalities of the computing apparatus 120 as described may also be integrated to an entity configured also to perform other operations. The specific examples provided in the description given above should not be construed as limiting the applicability and / or the interpretation of the appended claims. Lists and groups of examples provided in the description given above are not exhaustive unless otherwise explicitly stated.

Claims

WHAT IS CLAIMED IS:

1. A method for evaluating a condition of an elevator (100), the method, performed by a computing apparatus (120), comprises: receiving (210) a number of indicators indicative of at least one state occurring in the elevator (100), generating (220) a fingerprint by aggregating the number of indicators, the fingerprint being descriptive of the condition of the elevator (100) at an instant of time, determining (230) one or more matches in a predefined accuracy between the fingerprint descriptive of the condition of the elevator (100) generated at the instant of time and at least one other fingerprint descriptive on a condition of an elevator (100) at at least one other instant of time, generating (240), in response to determining the one or more matches in the predefined accuracy, an output comprising data descriptive of a prediction of a future condition of the elevator (100) based on history data descriptive of a development of a condition of the elevator (100) whose fingerprint generated the match with the generated fingerprint.

2. The method according to claim 1 , wherein the number of indicators forming the fingerprint is a predefined set of state values obtainable from the elevator.

3. The method according to claim 1 or claim 2, wherein the fingerprint is a heatmap representation.

4. The method according to any of the preceding claims, wherein the determination of the one or more matches is performed with a machinelearning model trained for the task.

5. The method according to claim 4, wherein the machine-learning model is a transformer model.

6. The method according to claim 4 or claim 5, wherein the aggregation of the number of indicators is performed with an embedding operation.

7. The method according to any of the preceding claims, wherein the generation of the output comprising data descriptive of the prediction of the future condition of the elevator (100) is performed based on at least one of the following: at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the same elevator (100) to that fingerprint that generated the one or more matches with the generated fingerprint; metadata associated with at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the same elevator (100) to that fingerprint that generated the one or more matches with the generated fingerprint wherein the metadata being descriptive of a condition of the elevator (100) at an instant the respective fingerprint is generated.

8. A computing apparatus (120) for evaluating a condition of an elevator (100), the computing apparatus (120) is configured to perform: receive (210) a number of indicators indicative of at least one state occurring in the elevator (100), generate (220) a fingerprint by aggregating the number of indicators, the fingerprint being descriptive of the condition of the elevator (100) at an instant of time, determine (230) one or more matches in a predefined accuracy between the fingerprint descriptive of the condition of the elevator (100) generated at the instant of time and at least one other fingerprint descriptive on a condition of an elevator (100) at at least one other instant of time, generate (240), in response to determining the one or more matches in the predefined accuracy, an output comprising data descriptive of a prediction of a future condition of the elevator (100) based on history data descriptive of adevelopment of a condition of the elevator (100) whose fingerprint generated the match with the generated fingerprint.

9. The computing apparatus (120) according to claim 8, wherein the computing apparatus (120) is configured to apply a predefined set of state values obtainable from the elevator as the number of indicators forming the fingerprint.

10. The computing apparatus (120) according to claim 8 or claim 9, wherein the computing apparatus (100) is configured to generate a heatmap representation as the fingerprint.11 . The computing apparatus (120) according to any of the preceding claims 8 to 10, wherein the computing apparatus (120) is configured to perform the determination of the one or more matches with a machine-learning model trained for the task.

12. The computing apparatus (120) according to claim 11 , wherein the machine-learning model is a transformer model executed by the computing apparatus (120).

13. The computing apparatus (120) according to claim 11 or claim 12, wherein the computing apparatus (120) is configured to perform the aggregation of the number of indicators with an embedding operation.

14. The computing apparatus (120) according to any of the preceding claims 8 to 13, wherein the computing apparatus (120) is configured to perform the generation of the output comprising data descriptive of the prediction of the future condition of the elevator (100) based on at least one of the following: at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the same elevator (100) to that fingerprint that generated the one or more matches with the generated fingerprint; metadata associated with at least one consecutive fingerprint to the fingerprint that generated the one or more matches wherein the at least one consecutive fingerprint is from the sameelevator (100) to that fingerprint that generated the one or more matches with the generated fingerprint wherein the metadata being descriptive of a condition of the elevator (100) at an instant the respective fingerprint is generated.

15. A computer program comprising instructions to cause the computing apparatus (120) of claim 8 to execute the steps of the method of claim 1 .

Images