Analyzing fault diagnosis reports for lighting fixtures

Abstract

The present disclosure relates to a system and method for analyzing and / or processing fault reports for lighting fixtures, and more specifically for determining the outcome of diagnosis of reported faults associated with lighting fixtures. The method comprises obtaining a fault report data comprising at least a first type of information including first data related to an identity of a lighting device. The lighting device may be arranged in a target space. A fault may have been reported for the lighting device in a fault report created by a user. The method further comprises extracting, by the processing device, one or more identity markers for forming the identity of the lighting device through parsing the first data comprised in the obtained fault report data. Further, the method comprises determining an initial identity of the lighting device based on the extracted one or more identity markers.

Description

[0001] 2024PF80539

[0002] 1

[0003] ANALYZING FAULT DIAGNOSIS REPORTS FOR LIGHTING FIXTURES

[0004] FIELD OF THE INVENTION

[0005] The present invention generally relates to a method and a system of determining an identity of a lighting device arranged in a space, such as a room and more specifically to determining outcome of a reported fault for the identified lighting device.

[0006] BACKGROUND OF THE INVENTION

[0007] One of the main factors playing an important role in customer satisfaction for connected lighting devices, otherwise referred to as smart lighting devices, is the effectiveness of the solutions that are provided to a user of a faulty lighting device in an event of contacting a customer care center. Usually, when a fault or malfunctioning of a lighting device is recognized by the user, a customer care agent is contacted by the user for assistance.

[0008] Typically, companies provide communication channels for the users to reach the customer service that may include social media channels or dedicated live chat platforms or smartphone fault reporting applications. When the user contacts the customer service through their devices on one of these platforms, options are provided for the user to provide information and necessary details regarding the defect or performance issue associated with the lighting device to the agent. The customer care agent in turn may guide the user through troubleshooting steps, asking further questions about the fault, and assist the user in solving the issue remotely.

[0009] However, often these platforms lack real-time communication features, e.g., phone call, online call or video call options. Furthermore, since the customer service may be global, and each agent may need to serve multiple users simultaneously, the correspondence and the provided responses to each may not be instant. It is also very common that the user may need to exit the live chat platform to perform certain actions provided in the solution by the agent, and simply not return to the chat anymore. In any case, the effective flow of communication between the user and the agent may be interrupted, leaving the user frustrated and the outcome of the reported issue unclear. This problem is reflected in the statistics showing that only a small fraction of users reporting performance issues on these platforms2024PF80539

[0010] 2

[0011] tend to reply to the agent when the agent asks the user whether their issue has been resolved based on the provided assistance.

[0012] US2019320519A1 relates to a lighting fault diagnosis method including determining a fault item among a plurality of fault items based on fault symptom data of a test for each fault item, and recommending a repair method suitable for the determined fault item.

[0013] US11825583B1 relates to methods and systems for the management of the electric lighting circuits by monitoring one or more lighting circuits connected to a controller through one or more sensors.

[0014] US5769527A relates to a stage lighting system comprised of a plurality of lamp units which may have diverse communication protocols, functions and data parameters. The stage lighting system is controlled by a modular control system comprised of a modular controller mainframe interconnected with a plurality of control devices which may have diverse communications protocols and data formats.

[0015] TIAO TAN ET AL proposes a LLM-based Network Semantic Generation (LNSG) algorithm in “Adapting Network Information into Semantics for Generalizable and Plug and Play Multi -Scenario Network Diagnosis” (published on 22 March 2025). The algorithm integrates semanticization and symbolization methods to uniformly describe the entire multi-modal network information. Based on the LNSG and LLMs, it presents NetSemantic, a plug-and-play, data-independent, network information semantic fault diagnosis framework.

[0016] US2024346458A1 discloses a method of automatically generating, by the one or more processors using a generative Al model, a structured service report in the predetermined format from an unstructured service report corresponding to a service request handled by one or more technicians for servicing a building equipment.

[0017] CN118333635 A relates to a method and system for achieving automatic fault reporting based on a large language model.

[0018] US2002174380A1 relates to a remote central helpdesk for a plurality of POS appliances includes a diagnostics engine which, on input of a reported problem, executes a series of manual and automated queries in a sequence determined by a decision tree and by the answers to the queries.

[0019] Accordingly, there is a need in the field of resolving fault diagnoses for connected lighting devices for effective solutions to determine whether an issue associated2024PF80539

[0020] 3

[0021] with a lighting device has been successfully resolved in order to make the fault reporting process more efficient and increase customer satisfaction.

[0022] SUMMARY OF THE INVENTION

[0023] It is an object of the present invention to provide a method and a system of analyzing and / or processing user generated fault reports, determining an identity of the lighting device in the fault report as well as determining outcome of diagnosis of the reported fault associated with the identified lighting device as defined in the appended claims herein. Various embodiments of the present invention have been defined in the dependent claims.

[0024] The term exemplary is in the following to be interpreted as serving as an example, example implementation, instance, scenario or illustration.

[0025] According to a first aspect of the present invention, there is provided a computer-implemented method for analyzing and / or processing fault reports for lighting fixtures, and more specifically for determining the outcome of diagnosis of reported faults associated with identified lighting fixtures. The method comprises obtaining, by a processing device, a fault report data comprising at least a first type of information including first data related to an identity of a lighting device. The lighting device is arranged in a target space. In several embodiments, a fault has been reported for the lighting device in a fault report created by a user. The method further comprises extracting, by the processing device, one or more identity markers for forming the identity of the lighting device through parsing the first data comprised in the obtained fault report data. Further, the method comprises determining, by the processing device, an initial identity of the lighting device based on the extracted one or more identity markers. The identity markers refer to those information or data which can be used to mark the identity of the lighting devices.

[0026] In some embodiments, the method may further comprise obtaining, by the processing device and from a lighting fixture database, log data associated with a plurality of lighting fixtures in communication with a server device. The log data may comprise recorded identity information of the plurality of lighting fixtures. Further, the method may comprise comparing the obtained log data to the determined initial identity of the lighting device in order to establish a match between the initial identity of the lighting device and a recorded identity information of a lighting fixture in the log data. If a match was determined by the comparison, the method may further comprise confirming and / or updating the determined initial identity of the lighting device based on the determined match in order to establish a unique identity of the lighting device.2024PF80539

[0027] 4

[0028] In some embodiments, the fault report information may further include second data related to the reported fault associated with the identified lighting device. Accordingly, the method may further comprise obtaining, by the processing device from a database, a plurality of predetermined fault classes associated with lighting fixtures. The method may further comprise extracting the reported fault for the identified lighting device through parsing the second data related to the reported fault and by comparing the parsed second data to the obtained plurality of predetermined fault classes. The method may further comprise identifying, between the plurality of the predetermined fault classes, a matching fault class indicative of a type of the extracted fault. Additionally, when the matching fault class is identified, the method may comprise classifying the extracted fault in the identified fault class.

[0029] In some embodiments, the method may further comprise providing a first input, by the processing device, to an artificial intelligence, Al, model configured to determine outcome of diagnosis of reported faults associated with lighting fixtures. The first input may comprise the fault report data. The method may comprise obtaining, by the processing device, a first output response from the Al model based on the provided first input, wherein the first output response comprises the initial identity of the lighting device determined by the Al model. The Al model may be configured to determine the initial identity by extracting the one or more identity markers for forming the identity of the identity of the lighting device through parsing the first data comprised in the obtained fault report data.

[0030] Further in some embodiments, the method may comprise providing, by the processing device, a first input to the Al model. The first input may comprise the obtained log data and the initial identity of the lighting device. In some examples, the initial identity of the lighting device may be determined by the Al model and provided in the obtained first output response of the Al model. The method may further comprise obtaining, by the processing device, an updated first output of the Al model comprising the established unique identity of the lighting device.

[0031] In several embodiments, the method may further comprise providing, by the processing device, the plurality of predetermined fault classes associated with lighting fixtures, and the established unique identity of the lighting device, as a second input to the Al model. In some examples, the established unique identity of the lighting device may be determined by the Al model and provided as a part of the second input. The method may further comprise extracting, by the Al model, the reported fault for the identified lighting2024PF80539

[0032] 5

[0033] device. Further, the method may comprise identifying, by the Al model and between the plurality of the predetermined fault classes, the matching fault class. In addition, the method may comprise classifying, by the Al model, the extracted fault in the identified fault class. Furthermore, the method may comprise obtaining, by the processing device, a second output response from the Al model based on the provided second input, wherein the second output response comprises a summary of the extracted fault and the identified fault class of the extracted fault.

[0034] In several embodiments, the method may further comprise determining, by the processing device and / or by the Al model, outcome of diagnosis of the reported fault associated with the identified lighting device. The method may thus comprise obtaining a current operational status data of the identified lighting device from the server device in communication with the identified lighting device. Further the method may comprise comparing the obtained current operational status data to the classified extracted fault, and determining if the reported fault has been successfully resolved based on said comparison.

[0035] In some embodiments presented herein, the Al model may be a multimodal large model (MM-LM) configured to analyze and / or process information from multiple data modalities. The method may thus further comprise providing a primary set of instructions to the MM-LM in order to contextually constrain the MM-LM to determine outcome of diagnosis of reported faults associated with lighting fixtures.

[0036] In several embodiments, the obtained fault report data may further comprise a second type of fault report information, wherein the first or the second type of fault report information include multiple data modalities being any one of textual data, image data or video data.

[0037] The present invention is based at least partly on the realization that by implementing an automated system, and utilizing deep product knowledge, the identity of faulty light fixtures and their associated performance issues can be determined based on usergenerated fault reports.

[0038] Moreover, the automated system greatly benefits from utilizing an Al model, e.g., a multimodal large model in combination with predefined issue categories built based on the product knowledge to further enhance the efficiency of customer support operations.

[0039] Furthermore, certain checks or actions may be performed automatically to help the customer care team determine whether the user’s issue for the identified lighting device has been resolved or not. Outcome of the automated determinations and steps of the2024PF80539

[0040] 6

[0041] presented technology also assists in providing further support to the users and enhances the process to alleviate the performance issues.

[0042] For instance, when the respective outcomes of the reported faults associated with the identified lighting devices are determined, the customer care team may proactively contact those users whose issues with the identified lighting devices remain unsolved to provide further support. This automated, efficient, accurate and swift customer care operation in turn leads to improved customer experience and satisfaction.

[0043] In the present context an Al model is to be construed a program that has been trained on a set of data to recognize certain patterns or make certain decisions without further human intervention. Foundation Al models may be pre-trained on large, unlabeled datasets and are capable of a wide array of applications. These versatile foundation models may utilize self-supervised learning during pre-training and can be fine-tuned for specific tasks or domains. A generative Al model may employ unsupervised or self-supervised learning to generate new data instances that mimic the statistical properties of the training dataset.

[0044] Generative Al models may be language models. A language model (LM) may be trained to predict the probability distribution of word sequences, facilitating various natural language processing tasks. Large language models (LLMs), exemplified by architectures like GPT-3 and GPT-4, leverage transformer-based deep learning techniques and are trained on extensive corpora of text to achieve advanced language understanding and generation capabilities.

[0045] Generative Al models may be multi-modal models designed to process and integrate multiple types of data inputs (modalities) such as text, images, audio, and video, and generate outputs that can also span these different modalities.

[0046] Multi-modal large models (MM-LM) are more advanced Al systems designed to process and understand multiple types of data modalities to perform complex tasks, leveraging the strengths of each modality to enhance overall performance and accuracy. MM-LMs may be considered an extension of LLMs with the added capability to process and understand other data types beyond text. An MM-LM may be built upon the foundational architecture of an LLM and incorporate additional modules or mechanisms to handle various data modalities such as images, audio, and video.

[0047] According to a second aspect, there is provided a computer program product comprising instructions which, when the program is executed by one or more processors of the processing device, causes the processing device to carry out the method according to any one of the embodiments of the method disclosed herein.2024PF80539

[0048] 7

[0049] According to a third aspect of the present invention, there is provided a system comprising a processing device configured to obtain a fault report data comprising at least a first type of information including first data related to an identity of a lighting device. The lighting device is arranged in a target space. A fault has been reported for the lighting device in a fault report created by a user. The processing device is further configured to extract one or more identity markers for forming the identity of the lighting device through parsing the first data comprised in the obtained fault report data. Further, the processing device is configured to determine an initial identity of the lighting device based on the extracted one or more identity markers.

[0050] Further embodiments of the different aspects are defined in the dependent claims. It is to be noted that all the embodiments, elements, features and advantages associated with the first aspect also analogously apply to the second, and third aspects of the present disclosure and vice versa.

[0051] These and other features and advantages of the present disclosure will be further explained in the following detailed description.

[0052] BRIEF DESCRIPTION OF THE DRAWINGS

[0053] This and other aspects of the present invention will now be described in more detail, with reference to the appended drawings showing embodiments of the present invention. The drawings are only schematic and the relative dimensions of some structures and layers may be exaggerated and not drawn to scale. Rather the dimensions may be adapted for illustrational clarity and to facilitate understanding.

[0054] Figs, la - b show schematic illustrations of a fault report interface comprising information pertaining to faulty lighting devices according to some embodiments of the present disclosure.

[0055] Figs. 2a - b show schematic block diagram illustrations of a system implementing a method according to several embodiments of the present disclosure.

[0056] Fig. 3 shows a schematic flowchart illustrating a method in accordance with several embodiments of the present disclosure.

[0057] DETAILED DESCRIPTION

[0058] The following disclosure presents various embodiments or examples for implementing different aspects of the provided subject matter. Specific examples of implementing the method are described to simplify the disclosure. These examples are not2024PF80539

[0059] 8

[0060] intended to be limiting. For instance, the description of a scenario of a reported fault for a particular lighting device and / or a particular space such as a bedroom may be equally applicable to another space such as a living room or other variations of the same space having different space-related attributes or other types of lighting devices. Additionally, reference numerals and / or letters may be repeated in various examples for simplicity and clarity, without implying a specific or limiting relationship between the different embodiments and / or configurations discussed, unless stated otherwise.

[0061] It is further to be noted that terms such as “first” and “second” etc. with reference to elements or steps may be used herein as labels to facilitate distinguishing between different elements and need not necessarily imply that such elements or steps are arranged or performed in that particular order, unless stated otherwise.

[0062] Those skilled in the art will appreciate that the steps, services and functions explained herein may be implemented using individual hardware circuitry, using software functioning in conjunction with one or more programmed microprocessors, microcontrollers, or general-purpose computers, using one or more Application Specific Integrated Circuits (ASICs) and / or using one or more Digital Signal Processors (DSPs) or some other programmable logical device, such as a Field Programmable Gate Array (FPGA). It will also be appreciated that when the present disclosure is described in terms of a method, it may also be embodied in one or more processors and one or more memories coupled to the one or more processors, wherein the one or more memories store one or more programs that perform the steps, services and functions disclosed herein when executed by the one or more processors.

[0063] Fig. 1 shows a schematic illustration of an interface 1 through which a user of one or more lighting devices 110 may attempt reporting a performance issue of a faulty lighting device 110 to a customer care service. The interface 1 in several embodiments may be a mobile phone or a handheld device of the user such as a personal computer, laptop, smart tablet or the like. The interface 1 may have a display allowing the user to interact and communicate with a software or an application for fault reporting or communication with the customer care service. The software or the application may be a dedicated application such as a lighting control application for smartphones, or a dedicated portal or function of the lighting control application for reporting performance issues to the manufacturer of the lighting devices. The lighting control application and the one or more connected lighting devices may be comprised in or connected to a smart home system or a connected network of smart devices.2024PF80539

[0064] 9

[0065] The plurality of smart devices and the lighting control application may be connected to each other over a communication network such as a home network through wired and / or wireless communication technologies such as wired local area network (LAN) or Ethernet, or any wireless link such as wireless LAN, Wi-Fi, Bluetooth, etc.

[0066] The application may allow the user to input different types and formats of information including textual, visual and / or audio data. The visual information may include images and / or video files uploaded to the fault report portal. The textual, visual or audio information may allow the user to explain details pertaining to an identified fault or performance issue for a lighting device 110 arranged in a space 100 (also referred to as target space 100) such as a room, e.g., a bedroom or a living room or any other indoor or outdoor space.

[0067] As recognized by the present inventors, when users typically contact customer service through such integrated text chat and email functions, users use text, images and videos to chat with customer service agents. However, absence of real-time communication features and instances that users may exit the fault report application or portal to perform certain actions often result in lack of insight to determine the outcome of a diagnosis and assistance provided for a reported fault. Thus, embodiments described herein provide simple and straight forward solutions in order to identify the faulty lighting devices and trace the reported issue in order to establish if the reported problems with the identified lighting devices have been resolved. Accordingly, relevant actionable steps may be followed by the customer care agents based on the determined outcome of the fault report diagnosis. For example, when the outcome of the reported faults associated with the identified lighting device is determined, the customer care team may contact those users whose issues with the identified lighting devices remain unsolved to provide further support. Details of the solutions provided herein will be described in the following with reference to exemplary implementations and scenarios to elucidate the aspect and embodiments of the invention.

[0068] In the reset of this description, examples may be discussed with respect to a certain lighting device arranged in a particular space, such as a ceiling lamp 110 arranged in a bedroom 100 as shown in Fig. lb. However, it should be clear that all the discussions, features, and advantages described with respect to a single space or lighting device analogously apply to a multitude of spaces or sub-spaces within a space comprising a plurality of lighting devices without departing from the scope of the present claims.

[0069] The example space 100 may comprise one or more controllable or connected lighting devices 110. It should be appreciated that the one or more controllable lighting2024PF80539

[0070] 10

[0071] devices may be of any other type, shape or function, e.g., they may be wall-mounted luminaires, floor lamps, tabletop lighting arrangements or the like. The controllable or connected lighting devices may be simply referred to as “lighting devices” or “lighting fixtures” in the rest of this description for the sake of brevity.

[0072] By a controllable or connected lighting device herein it is meant a lighting device configured to be connected to a control system or external hardware device comprising control software for steering operation of the lighting device. For instance, the connected lighting devices may be connected to a user’s smartphone with a lighting control application wirelessly controlling operation of the connected lighting device. The lighting device may produce controllable lightning properties such as intensity of the emitted light, temperature of the emitted light, color or hue of the emitted light, etc. The operation, settings or properties of the controllable lighting device may be controlled and manipulated through various control mechanisms. These mechanisms may include non-limiting examples of operational controls to turn on and / or turn off the light, dimming controls to adjust the intensity of the light output, or color temperature controls to change the color temperature of the light ranging from warm (e.g., yellowish) to cool (e.g., bluish) tones.

[0073] The lighting devices may generate light through different mechanisms which are per se known in the art. For example, the lighting devices may comprise light emitting diodes (LED), halogen lamps, fluorescent lamps, light bulbs, filaments, etc. to create a wide spectrum of lighting properties.

[0074] The one or more controllable lighting devices may comprise electronic equipment such as integrated sensor devices, wired and / or wireless communication interfaces, processing circuitry, and the like to operate the lighting device and connect to the external control devices.

[0075] The lighting device in Fig. lb is a ceiling lamp 110 arranged in a bedroom and Fig. la includes a portion of a fault report generated by the user to report performance issues with the ceiling lamp 110.

[0076] As shown in the example of Fig. la, the interface 1 used by the user is a smartphone device 1, through which the user has launched a lighting control application for reporting performance issues with the example lighting device 110. The user has initiated a chat sequence with the customer service agent to notify the agent of one or more particular performance failures with the lighting device 110. In thread 11 the user briefly describes the issue they are experiencing and receives a response from the customer care agent in thread 12. The correspondence continues in thread 13, wherein more detailed information regarding2024PF80539

[0077] 11

[0078] the specific lighting device and the relevant performance issue is shared with the agent, and possible assistance to resolve the issue is received by the user. Several types of performance issues may be reported for a lighting device that may include one or more of non-exhaustive issues related to lighting attributes or functionality problems such as flickering issues, dimming issues, color temperature anomalies or malfunctioning, noise problems, or disruption in one or more control functions such as on / off function or light intensity control. In the example of Figs, la - lb, the reported issue for the ceiling lamp 110 is light flickering and inability to control the lighting device, as partly indicated by the user in thread 11. As seen in thread 13, more details regarding the faulty lighting device 110 are provided by the user. For instance, a screenshot of the light control application depicting information related to the faulty light 110 is provided to the agent. The information in this example relate to lighting device’s identity 13a, 13c including a name or alias or an IP address, its lighting attributes 13b such as color temperature or light intensity or the like, and an error message 13d appearing on the control application (screenshot image data). Furthermore, the user has uploaded an image 13e of the flickering light 110 to the application. The image 13e may have been captured by a smartphone camera through an external camera application or one directly embedded in the fault report application accessing the smartphone camera in order to capture real-time image or video data to be transmitted to the customer care agent.

[0079] It is also shown that the agent has asked the user in thread 14 whether their issue has been resolved after the correspondence and received assistance, however, the user has not responded to this request. Therefore, the outcome of the fault diagnosis for the reported fault is unclear to the customer care agent. According to several embodiments herein, there is provided a system and a method for determining the outcome of diagnosis of reported faults associated with lighting fixtures such as the example lighting device 110 based on fault reports created by users of lighting fixtures.

[0080] Fig. 2a shows a schematic block diagram of an exemplifying system 200 which may be used to implement a method 300 of the present disclosure. The system 200 comprises a processing device 210 to carry out the steps of embodiments of method 300. The processing device 210 may comprise one or more processors 210a, such as CPUs or GPUs, or any other suitable microprocessors, microcontrollers, ASICs, FPGAs, one or more memory or data storage modules 210b and the like. The processing device 210 may comprise additional modules such as receiver modules (not specifically shown) for receiving input data including fault report data, analysis modules (not specifically shown) configured to analyze and process the received input data and the like. The processors 210a or processing circuity2024PF80539

[0081] 12

[0082] in the processing device 210 may be configured to carry out the several functions and operations of the presented technology herein.

[0083] In some embodiments, the system 200 may be connected to external networks such as a cloud network (not shown) via wireless links in order to access a remote server comprising processing capabilities, memory devise, stored databases, and the like.

[0084] The collected data such as fault report data, as will be discussed further below, may be stored in the memory modules 210b of the processing device 210 or be transmitted and stored in memory modules of the remote server such as the server device 230 shown in Fig. 2a for later use.

[0085] Accordingly, the processing device 210 is configured to obtain a fault report data generated based on the user fault report. The fault report data comprises at least a first type of information including first data related to an identity of a lighting device 110 for which a fault has been reported in the fault report created by the user. For instance, the threads 11 - 14 are included in the fault report data, collectively referenced as fault report data 10. In some embodiments, the first data may be unstructured data. The unstructured data herein means data prior to being analyzed and parsed into an organized or classified format and category. The unstructured data in the user report comprises information related to the identity of the faulty lighting device that will be analyzed and processed in order to determine the identity of the lighting device having performance issues. The identity of the lighting device is a unique identifier for that lighting device to recognize and distinguish that lighting device, which is either set by the manufacturer (default mode) or assigned by the user e.g., when initializing the lighting device and connecting it to the lighting control application or the home network.

[0086] The fault report data 10 is provided to the processing device 210 as shown in the example of Fig. 2a. The processing device 210 is configured to extract one or more identity markers 13a - 13c of the lighting device 110 through parsing the first data, e.g., unstructured data comprised in the obtained fault report data 10. The one or more extracted identity markers will be utilized for forming the identity of the lighting device. In other words, the identity of the lighting devices will be discerned based on the one or more extracted identity markers.

[0087] In some embodiments, the obtained fault report data may further comprise a second type of fault report information. The first or the second type of fault report information may include multiple data modalities. The data modalities may be any one of textual data, image data or video data or a combination thereof.2024PF80539

[0088] 13

[0089] The user may specify and assign customized names or aliases, e.g., ceiling lamp (here the exemplary name representing the type of the lighting device being a ceiling lamp) or installation locations, e.g., bedroom, and the like.

[0090] Accordingly, the one or more identity markers such as example identity markers 13a - 13c or 13f, 13g of the lighting device shown in Figs, la and lb may include type of the lighting device, a particular space or location 100 at which the lighting device is arranged, an installation position of the lighting device in the target space 100, an alias of the lighting device assigned to the lighting device by the user or the control application or the manufacturer, color temperature range, light intensity, color, the IP address, or any other identity markers indicting unique product information such as product serial number, to name a few.

[0091] The plurality of the identifying markers as defined herein may collectively form the identity profile of the faulty lighting device that are included within the unstructured textual and / or visual data of the user-created fault report. The lighting device may be identified either by one or a combination of its identity markers.

[0092] By parsing the unstructured data, the processing device 210 extracts the identity markers of the lighting device and determines an initial identity of the lighting device based on the extracted one or more identity markers. In some examples, the determined initial identity of the faulty lighting device may comprise its complete list of identity markers presented in a particular format such as a JavaScript object notation (JSON) format. In some embodiments however, the determined initial identity may only be represented by some selected identity markers such as the alias, device type or class, and its location structured in the intended data format such as JSON.

[0093] For instance, the processing device 210 may be configured to parse the fault report data 10 in Fig. la and extract several attributes and identity markers 13a - 13c associated with the faulty lighting device 110. At least one of the extracted identity markers may be arranged in a data format such as the example JSON format below to stand for the initial identity of the faulty lighting device 110.

[0094] {"location" : [Parker ’s room], “light alias”: “A 19 Warm white 2 ”}

[0095] In some embodiments, the determined initial identity may include the complete list of identity markers of the faulty lighting device 110, for instance:2024PF80539

[0096] 14

[0097] {“id”:l, “type”: “ceiling lamp”, “location” : [Parker’s room], “light alias” : “A19 Warm white 2”, “IP address”: “192.16.2.1” “color temperature” : “warm white” ; “3000 K”, “Light intensity ”: “0 ”}

[0098] It is clearly conceivable that the determined initial identity of the lighting device could be organized in any other suitable data interchange format such as a tabular text format like comma-separated values (CSV) or the like.

[0099] It should be appreciated that the processing device 210 may use any suitable algorithm, hardware or software program such as text or speech recognition algorithms, image processing algorithms or neural network models for purposes of implementing various embodiments and aspects of the present invention. These algorithms may be used to parse multimodal information, accessing and retrieving data from databases, performing semantic searches in databases or any other related functions. Basic working principles of such models and algorithms are presumed to be available to a skilled person in the art and will not be discussed any further herein.

[0100] In some embodiments, for instance as shown in the example system 200 of Fig. 2a, the processing device 210 may be configured to comprise or be in communication with an artificial intelligence (Al) model 400.

[0101] By being in communication with the Al model 400 is to be construed that the processing device 210 is directly or indirectly, e.g., via intermediate entities connected to the Al model 400 and can transmit and / or receive data, information or command signals to and from the Al model 400. In some embodiments the system 200 may further comprise the Al model 400, i.e., as a component or entity of the system 200. For instance, the Al model 400 may at least partly be implemented by the hardware and software of the processing device 210. In some example implementations, the Al model 400 may be a standalone solution implemented in dedicated hardware and software while being locally integrated with the system 200. In some examples, the Al model may be implemented remotely e.g., at the remote server 230 comprised in a cloud-based solution and configured to be in communication with the system 200. Accordingly, in some examples, generated inputs or prompts intended for the Al model 400 may be transmitted by the processing device 210 to the remote server 230 to be inputted to the Al model 400.

[0102] To this end, the Al model 400 may be implemented using one or more CPUs, DSPs, GPUs, memory modules, or any other suitable microprocessors, microcontrollers,2024PF80539

[0103] 15

[0104] ASICs, FPGAs, or a combination of the mentioned discrete analog and / or digital components.

[0105] In some embodiments the processing device 210 may be in communication with the Al 400 through intermediate entities or interfaces such as an application programming interface (API) (not specifically shown) for transmitting and / or obtaining data, instructions, prompts, responses or the like through API calling processes.

[0106] In several embodiments, the processing device 210 may be configured to provide a first input 401, to the Al model 400 configured to determine outcome of diagnosis of reported faults associated with lighting fixtures. The first input 401 may at least comprise the fault report data. The processing circuitry may be configured to obtain a first output response 404a from the Al model based on the provided first input 401, wherein the first output response 404a may comprise the initial identity of the lighting device determined by the Al model 400. Thus, in some example implementations, the processing device 210 may extract the one or more identity markers by utilizing the Al model 400. Accordingly, the Al model 400 may be configured to determine the initial identity by extracting the one or more identity markers for forming the identity of the identity of the lighting device through parsing the first data comprised in the obtained fault report data.

[0107] The Al model 400 may be a generative-AI model 400. In various embodiments, the generative-AI model may be a multimodal model such as a multimodal large model (MM-LM) 400.

[0108] In some embodiments, the MM-LM may be a large language model (LLM) 400 that is configured to receive various types of data such as text, image, video or audio data. In some embodiments, the MM-LM may be a general -purpose model. In some example implementations the MM-LM 400 may be a general-purpose model that is pretrained or finetuned with a domain-specific dataset obtained and prepared for the domain of analyzing and / or processing fault reports for lighting fixtures. The specialized training dataset may comprise image data, text data, video data collected over time from a plurality of fault reports gathered for a large number of different scenarios and issues. In some embodiments, databases containing predefined fault classes, as will be discussed further below, may be built based on the continuously collected fault report data. The databases may be subject to ongoing or periodic updates based on the data collected.

[0109] Accordingly, the system 200 may be configured to provide a primary set of instructions to the MM-LM 400 in order to contextually constrain the MM-LM 400 to analyze and / or process customer reports for lighting fixtures, and more specifically to2024PF80539

[0110] 16

[0111] determine outcome of diagnosis of reported faults associated with lighting fixtures. By primary set of instructions, it is meant an initial set of instructions to contextually constrain the MM-LM 400 to its specific domain of operation.

[0112] To contextually constrain the MM-LM 400 herein is to be understood as prompting, preparing or conditioning the MM-LM 400 by means of initial or primary instructions and / or parameters and thereby defining the domain of operation, context, as well as roles and responsibilities of the MM-LM 400. Thus, contextual constraints may be applied to the MM-LM 400 through a set of prompts including specific instructions or examples of domain-specific tasks, system instructions, or other methods such as fine-tuning in order to guide the output responses of the MM-LM 400 for the specific task and domain of analyzing and / or processing fault reports for lighting fixtures, and more specifically to the domain of determining the outcome of diagnosis of reported faults associated with lighting fixtures. Optionally, the primary set of instructions may comprise a predefined structure of the output response of the MM-LM, specific data formats such as JSON to structure the output response, examples of such data formats and the like.

[0113] In some embodiments the MM-LM model 400 may be prompted in advance of receiving the first 401 input to contextually constrain the MM-LM 400 to its specific domain of operation. In some embodiments however, the primary set of instructions may be comprised in the first input 401 and be provided to the MM-LM 400.

[0114] In some embodiments, the processing device 210 is further configured to obtain log data associated with a plurality of lighting fixtures in communication with the server device 230 from a lighting fixture database 230a. The log data and the lighting fixture database 230a may comprise the user information as well as unique product information of all the registered connected lighting devices that have been or currently are in operation. The log data may also comprise the identity information of all lighting devices registered to the specific user who has created the fault report. In some example embodiments, the log data may be included in the backend data typically stored in databases and including user information, content, settings, and other essential data required for the lighting control application’s functionality.

[0115] Accordingly, log data may comprise recorded identity information of the plurality of lighting fixtures, including the lighting device 110 for which the fault report has been generated. In some embodiments, the server device 230 may be in communication with the connected lighting devices through the lighting control application.2024PF80539

[0116] 17

[0117] The processing device 210 may be configured to compare the obtained log data to the determined initial identity of the faulty lighting device 110 in order to establish a match between the initial identity of the lighting device 110 and a recorded identity information of a corresponding lighting fixture in the log data. This way the processing device 210 would reconcile the initial identity of the faulty lighting device with backend data identifiers. Thus, the processing device 210 may ensure that the lighting device has been identified correctly and thus confirm the determined initial identity of the lighting device 110. However, if the comparison reveals that the initial identity should be rectified, the processing device 210 may accordingly update or correct the initial identity based on the retrieved log data having the accurate identity information. In either case, the processing device 210 is configured to establish a unique identity of the lighting device 110. This final identity will be used for additional processing steps in order to analyze and process the outcome of the reported faults for the identified lighting device 110.

[0118] In some embodiments, as shown in the example Fig. 2b, the processing device 210 may further be configured to provide a first input 401 to the Al model 400, wherein the first input may comprise the obtained log data 403 from the database 230a and the determined initial identity 404a’ of the lighting device 110. In some example embodiments, the initial identity 404a’ of the lighting device 110 may be determined by the Al model 400 and be provided by the obtained first output response 404a of the Al model. The processing device 210 may be further configured to obtain an updated first output 404a” of the Al model 400 based on the provided log data 403 and initial identity 404a’ inputs, wherein the first updated output 404a” comprises the established unique identity of the lighting device 110.

[0119] In the example scenario shown in Fig. lb, the fault report data may include a time-lapse image acquisition or a captured video file from the faulty lighting device 110 (ceiling lamp) arranged in the bedroom 100. In this example only three sequential frames 21 - 23 of the video or time lapse image data are depicted to explain a simplified scenario. For the sake of simplicity this example refers to the same ceiling lamp 110, however the lamp 110 is installed in a different location than the discussed “Parker’s room” in Fig. la. It should be clear that the video file may have any suitable size, or as many suitable frames of the video data may be used to be parsed by the processing device 210. It is also conceivable that any suitable predefined data size limits may be embedded into the fault reporting application, constraining the size of images or video files that could be uploaded by the user into the fault report. The recorded frame 21 shows that the lighting device 110 is illuminated at normal intensity “A”, but the frame 22 shows that the light is turned off entirely and is in the shut-off2024PF80539

[0120] 18

[0121] state “B”, while frame 23 shows that the light is blinking with an attenuated intensity level “C”. Hence, a flickering performance issue is noticed, and reported by the user by means of the captured video or time lapse data. The time lapse or video data may be provided independently in the fault report. Alternatively, the video or time lapse image data may be provided in addition to the textual or single-shot image data as e.g., explained with reference to Fig. la.

[0122] In example scenarios when the user has not provided adequate information including enough details in the first data for extracting the identity markers of the faulty lighting device 110, the embodiments of the technology presented herein that obtain log data and establish the unique identity of the faulty lighting device are particularly advantageous. For instance, if the user only provided the video recording 13h of a flickering lighting device 110 in their bedroom 100 as shown in Fig. lb without any other information describing the location or type of the lighting device, the processing device 210 may not be able to immediately distinguish the identity of the lighting device from the fault report data. The processing device 210 may parse the video data and determine that the space 100 could be a bedroom (as a first exemplary identity marker 13f) and that the lighting device may be a ceiling light 110 (type or alias as an exemplary second identity marker 13g) having a performance issue of flickering, thus creating an initial identity such as:

[0123] {"location ": [bedroom ], “light alias “ceiling light ”}

[0124] As such, an additional step may be required in order to establish the final identity of the lighting device 110. The processing device 210 may be configured to access the server device 230 to retrieve relevant log data in the lighting fixture database 230a in order to find a lighting device that matches the extracted identity markers "bedroom" 13f and "ceiling light” 13g. In order to narrow down the search, the search in the database may be more personalized in order to obtain the log data of all the lighting devices registered to the particular user who has created the fault report. A list of all lighting devices arranged in all of the rooms in the user’s home and their respective identity profiles may be retrieved from the lighting fixture database that could look like an example retrieved information of user’s lighting devices in the list below.

[0125] Bathroom: light 1, light 2

[0126] Bedroom: Floor Lamp, Desk lamp2024PF80539

[0127] 19

[0128] Dining: Dinning Light 1, Dinning Light 2, Downlight

[0129] Kitchen: Al 93

[0130] Boys Room: Floor Lamp

[0131] Girls Bedroom: Ceiling

[0132] Living room: Ceiling Fans, Blue Lamp

[0133] Accordingly, the processing device 210 may be configured to perform a comparison between the initially extracted identity markers form the video data and the retrieved list of identity profiles in order to confirm and / or update the initial identity of the faulty lighting device 110. In this case, the processing device 210 may determine and update the initial identity to establish the final identity of the faulty lighting device 110 to be:

[0134] {"location ": [ Girls Bedroom ], “light alias ”: “Ceiling ”}

[0135] In several embodiments, the fault report information may further include second data, e.g., unstructured data related to the reported fault associated with the identified lighting device 110. In other words, the obtained fault report data will be further processed by the processing device to extract and associate the performance issue to the final identity of the identified lighting device 110. As mentioned in the examples of Figs, la and lb above, the identity of the ceiling lamp 110 may be tagged with the initial "flickering ' fault type. Furthermore, based on information extracted from the image data of the screenshot in thread 13d, the lighting device 110 in Parker’s Room may be initially tagged with “offline and cannot be controlled".

[0136] The processing device 210 may be further configured to obtain, from a database 230b, a plurality of predetermined fault classes associated with lighting fixtures. For example, from the database 230b is shown in Fig. 2a being comprised in the server device 230, wherein it will be accessible by the processing device 210 on demand. It should be clear that the databases 230a, 230b storing the identity information and the fault categories may be the same database, or they may be distributed databases, local databases or cloud-base databases, or a combination thereof to give a few non-limiting examples.

[0137] The obtained prior knowledge, i.e., classified issue types 30 comprising the plurality of predefined issue categories or fault classes CO - Cn (“n” denoting the number of example classes) may thus be provided to the processing device 210 in order to extract the reported fault for the identified lighting device 110. The plurality of the predefined fault2024PF80539

[0138] 20

[0139] classes CO - Cn may include any suitable number of categories, each denoting a particular type of issue and its attributes classified and labeled under a class name or alias representing the issue type. In a simplified example below, a list of some commonly occurring issues classified under respective fault classes is provided in a tabular format in Tabel. 1 to further elucidate the embodiments described above. Even though only details of seven exemplary issue classes are mentioned in Table. 1 it should be clear to a skilled person that any number of classes such as “n” classes denoting any other type of issues, having any other fault description than provided in this example are conceivable to be comprised in the list of plurality of fault classes CO - Cn.

[0140] Table 1

[0141] <>

[0142] <>

[0143] <> <>

[0144] <>

[0145] <>

[0146] <>

[0147] <> >

[0148]

[0149] The processing device 210 may thus be configured to parse through the second data related to the reported fault and compare the parsed second data to the obtained plurality of predetermined fault classes CO - Cn. Further, the processing device 210 may be configured to identify a matching fault class indicative of a type of the extracted fault. When a matching fault class is identified, the processing device will classify the extracted fault in the identified fault class. For example, the processing device 210 may determine and2024PF80539

[0150] 21

[0151] organize exemplary JSON lists below for the identified lighting device 110 in the example scenarios in Fig. la and lb respectively:

[0152] {"location": [Parker’s room], “light alias” : “A19 Warm white 2”, "Issue type": <Class 4> and <Class_7>]; and

[0153] {"location": [Girls Bedroom], “light alias” : “Ceiling”, "Issue type": <Class 7>}.

[0154] In some embodiments, as shown in the example of Fig. 2b, the processing device 210 may be further configured to provide the fault report data and the determined final identity of the faulty lighting device 110 as a second input to the Al model 400 to extract and classify the reported fault. More specifically, the processing device 210 may be configured to provide the classified issue types 30 including the plurality of predetermined fault classes CO - Cn associated with lighting fixtures, and the established unique identity 404a” of the lighting device, as the second input 402 to the Al model 400, e.g., the MM-LM 400. In some examples, the established unique identity of the lighting device may be determined by the Al model 400 and provided as a part of the second input 402. It should be appreciated that the first 401 and / or second 402 inputs may be provided to the Al model, e.g., the MM-LM 400 as a first 401 and a second 402 prompts generated by the processing device 210.

[0155] The Al model 400 may be configured to extract the reported fault for the identified lighting device 110. Further, the Al model 400 may be configured to identify between the plurality of the predetermined fault classes, the fault class matching the extracted fault. Moreover, the Al model 400 may be configured to classify the extracted fault in the identified fault class. The processing device 210 may be configured to obtain a second output response 404b from the Al model 400 based on the provided second input 402, wherein the second output response 404b comprises a summary of the extracted fault and the identified fault class of the extracted fault. In some embodiments, the second output 404b may comprise a specific data format such as the JSON format indicating the identity of the lighting device 110 and its classified performance issues.

[0156] For example, the processing device 210 based on the second output 404b of the Al model 400 may determine and organize exemplary JSON lists below, for the example scenarios in Fig. la and lb respectively:2024PF80539

[0157] 22

[0158] {"location": [Parker’s room], “light alias” : “A 19 Warm white 2”, "Issue type" : <Class 4> and <Class_7> , “issue summary”: “the A19 Warm white 2 is offline and cannot be controlled, and it is flickering”]; and

[0159] {"location" : [Girls Bedroom] , “light alias” : “Ceiling”, "Issue type": <Class 7>, “issue summary ”:

[0160]

[0161] When the relationship between the identified lighting device 110 and its associated fault(s) are established, the processing device 210 may be further configured to determine outcome of diagnosis of the reported fault associated with the identified lighting device 110. The processing device 21 may accordingly obtain a current operational status data of the identified lighting device from the backend, i.e., the server device 230 in communication with the identified lighting device 110. Further, the processing device 210 will compare the obtained current operational status data to the classified extracted fault and determine if the reported fault has been successfully resolved based on this comparison. In some embodiments, the Al model 400 may be configured to determine the outcome of the diagnosis by obtaining and comparing the current operational status of the identified lighting device to the classified extracted fault. In any case, the processing device 210 may utilize the Al model 400 for implementation of any one of the above functions and operations. The processing device 210 may utilize other established algorithms for the above comparison and determination operations. Additionally or alternatively, some of the steps may be performed by the Al model 400 and some of the operations by conventional algorithms.

[0162] In an example scenario, the operation of the lighting device, e.g., the lighting device having a fault in fault class <Class_4> may be monitored for a predetermined period of time such as 24 hours or the like and the operation status data be fetched from the server device 230 in arbitrary or predetermined intervals such as every 5 - 10 minutes. In case no indications, such as a user declaration of a resolved issue is found by the processing device 210 within the monitoring window, the processing device 210 may mark the identified performance issue as “unresolved” and generate a corresponding output signal 40a indicating this outcome. The output signal may be in the form of a notification 40 such as a visual, textual, or audible notification provided to the customer care agent. This notification may be utilized by the customer care center for taking optional further steps. In some embodiments, the notification may comprise backend information regarding the user such as their contact2024PF80539

[0163] 23

[0164] information in order for the agent to proactively contact the user for further follow ups and assistance on the status of the unresolved issue. In some cases, upon determining that the issue has in fact been solved, and no further action is needed, the notification signal 40 may be generated by the processing device 210 indicating this outcome.

[0165] Similarly, to establish whether the functionality of the flickering light 110 belonging to <Class_7> has been restored to normal, the operation status of the ceiling light may be retrieved in short periods of time, for example 5 times within 3 seconds, or similar time frames. The processing device will check if the dimming level and color temperature of the flickering light 110 show abnormal changes withing the retrieved time frame. If no change is detected, the issue will be marked “resolved”, however, if changes indicating persisting issues of flickering are identified the processing device may continue to monitor the operation of the lighting device 110 for a predefined period of time such as in 10-minutes intervals within a 24-hour time period. Based on the outcome this monitoring, if the performance issue still exists, the processing device 210 will generate a notification signal 40a to notify the customer care agent to take further actions such as proactively contacting the user. It should be appreciated that the above-provided scenarios are merely non-limiting examples of determining of processing user-generated fault report and determining the outcome of their diagnosis according to the advantageous method 300 performed by the exemplary system 200 and the processing device 210. Other scenarios are clearly conceivable. For example, the processing device 210 may continuously reiterate the comparison of the initial identity of the lighting device to the obtained log data to find a matching identity information in case initial searches in the database did not return a result. In case the processing device 210 or the Al model 400 did not find a match, an output of “nomatch-found” may be returned and the process may be terminated or reconfigured.

[0166] Accordingly, the presented methods and systems herein automatically and iteratively identify faulty lighting devices, their associated performance issues, and monitor the outcome of the reported fault. By providing such intelligence to the customer care team, a seamless customer support can be provided, wherein appropriate actions to resolve any remaining user issue are implemented promptly. This way, undesired delays and shortcomings in the customer service process are reduced and customer satisfaction metrics can be noticeably improved.

[0167] Fig. 3 shows a schematic flowchart of an exemplifying method 300 according to several embodiments of the present disclosure for analyzing and / or processing fault reports for lighting fixtures, and more specifically for determining the outcome of diagnosis of2024PF80539

[0168] 24

[0169] reported faults associated with identified lighting fixtures. In some embodiments, embodiments of the method 300 may be implemented by means of the system 200. The method 300 comprises obtaining 301, by a processing device 210, a fault report data comprising at least a first type of information including first data related to an identity of a lighting device 110. The lighting device 110 is arranged in a target space 100. In several embodiments, a fault has been reported for the lighting device 110 in a fault report 13 created by a user. The method further comprises extracting 303, by the processing device, one or more identity markers 13a - 13c, 13f, 13g for forming the identity of the lighting device through parsing the first data comprised in the obtained fault report data. Further, the method comprises determining 305, by the processing device, an initial identity of the lighting device based on the extracted one or more identity markers.

[0170] In some embodiments, the method 300 may further comprise obtaining 307, by the processing device and from a lighting fixture database 230a, log data associated with a plurality of lighting fixtures in communication with a server device 230. The log data comprises recorded identity information of the plurality of lighting fixtures. Further, the method may comprise comparing 309 the obtained log data to the determined initial identity of the lighting device in order to establish a match between the initial identity of the lighting device 110 and a recorded identity information of a lighting fixture in the log data. If a match was determined by the comparison, the method may further comprise confirming 311a and / or updating 311b the determined initial identity of the lighting device based on the determined match in order to establish 311 a unique identity of the lighting device.

[0171] In some embodiments, the fault report information may further include second data related to the reported fault associated with the identified lighting device 110.

[0172] Accordingly, the method 300 may further comprise obtaining 313, by the processing device from a database 230b, a plurality of predetermined fault classes CO - Cn associated with lighting fixtures. The method 300 may further comprise extracting 315 the reported fault for the identified lighting device 110 through parsing the second data related to the reported fault and by comparing 316 the parsed second data to the obtained plurality of predetermined fault classes. The method may further comprise identifying 317, between the plurality of the predetermined fault classes, a matching fault class indicative of a type of the extracted fault. Additionally, when the matching fault class is identified, the method may comprise classifying 319 the extracted fault in the identified fault class.

[0173] In some embodiments, the method 300 may further comprise providing 321a a first input 401, by the processing device 210, to an artificial intelligence, Al, model 4002024PF80539

[0174] 25

[0175] configured to determine outcome of diagnosis of reported faults associated with lighting fixtures. The first input 401 may comprise the fault report data. The method 300 may comprise obtaining 323a, by the processing device 210, a first output response 404a from the Al model 400 based on the provided first input, wherein the first output response comprises the initial identity of the lighting device 110 determined by the Al model.

[0176] Further in some embodiments, the method 300 may comprise providing 321b, by the processing device 210, a first input 401 to the Al model 400. The first input may comprise the obtained log data 403 and the initial identity of the lighting device 110. The initial identity of the lighting device 110 may in some examples be provided in the obtained first output response 404a’ of the Al model 400. The method may further comprise obtaining 323b, by the processing device 210, an updated first output 404a” of the Al model comprising the established unique identity of the lighting device 110.

[0177] In several embodiments, the method 300 may further comprise providing 321c, by the processing device 210, the plurality of predetermined fault classes CO - Cn associated with lighting fixtures, and the established unique identity of the lighting device, as a second input 402 to the Al model 400. In some examples, the established unique identity of the lighting device may be determined by the Al model 400 and provided as a part of the second input. The method 300 may further comprise extracting 315a, by the Al model, the reported fault for the identified lighting device. Further, the method may comprise identifying 317a, by the Al model 400 and between the plurality of the predetermined fault classes, the matching fault class. In addition, the method may comprise classifying 319a, by the Al model, the extracted fault in the identified fault class. Furthermore, the method may comprise obtaining 323c, by the processing device, a second output response 404b from the Al model based on the provided second input, wherein the second output response comprises a summary of the extracted fault and the identified fault class of the extracted fault.

[0178] In several embodiments, the method 300 may further comprise determining 325, by the processing device 210 and / or by the Al model 400, outcome of diagnosis of the reported fault associated with the identified lighting device. The method 300 may thus comprise obtaining 327 a current operational status data of the identified lighting device 110 from the server device 230 in communication with the identified lighting device. Further the method may comprise comparing 329 the obtained current operational status data to the classified extracted fault, and determining 331 if the reported fault has been successfully resolved based on said comparison.2024PF80539

[0179] 26

[0180] In some embodiments presented herein, the Al model may be a multimodal large model, MM-LM 400 configured to analyze and / or process information from multiple data modalities. The method may thus further comprise providing 321d a primary set of instructions to the MM-LM in order to contextually constrain the MM-LM to determine outcome of diagnosis of reported faults associated with lighting fixtures.

[0181] In several embodiments, the obtained fault report data may further comprise a second type 13e, 13h of fault report information, wherein the first or the second type of fault report information include multiple data modalities being any one of textual data, image data or video data.

[0182] In some embodiments, the method 300 may further comprise generating 333, by the processing device 210, an output signal 40a indicating the outcome of the fault report diagnosis. The output signal 40a may be in the form of a notification 40 such as a visual, textual, or audible notification provided to the customer care agent in order to take further actions accordingly.

[0183] The different features and steps of the embodiments may be combined in other combinations than those described. It should be noted that the word "comprising" does not exclude the presence of other elements or steps than those listed and the words "a" or "an" preceding an element do not exclude the presence of a plurality of such elements. It should further be noted that reference signs do not limit the scope of the claims, that the disclosure may be at least in part implemented by means of both hardware and software, and that several "modules" or "units" may be represented by the same item of hardware. As used herein, the term "if may be construed to mean "when or "upon" or "in response to determining or "in response to detecting" depending on the context. Similarly, the phrase "upon determining" may be construed to mean "when it is determined" or "in response to determining" or "upon detecting and identifying occurrence of an event" or "in response to detecting occurrence of an event" depending on the context. The term "obtaining" is herein to be interpreted broadly and encompasses receiving, retrieving, collecting, acquiring, and so forth directly and / or indirectly between two entities configured to be in communication with each other or with other external entities. Although the figures may show a specific order of method steps, the order of the steps may differ from what is depicted. In addition, two or more steps may be performed concurrently or with partial concurrence. It should also be appreciated that any of the embodiments and aspects discussed above may be carried out by utilizing the Al model, or any other appropriate algorithms in the art. As mentioned earlier, the processing device 210 may utilize established algorithms in the art such as image or text2024PF80539

[0184] 27

[0185] recognition and analysis algorithms for carrying out the functionalities disclosed herein. Additionally or alternatively, some of the steps may be performed by the Al model 400 and some of the operations by conventional algorithms.

[0186] Aspects of the present invention may be implemented in a computer program product comprising instructions which, when the program is executed by one or more processors of the processing device, causes the processing device to carry out the method according to any one of the embodiments of the method disclosed herein. Accordingly, there may be provided a computer-readable storage medium, such as a non-transitory computer-readable storage medium, storing one or more programs configured to be executed by one or more processors of the processing device. The one or more programs may comprise instructions for performing the method according to any one of the embodiments of the method of the present invention.

Claims

2024PF8053928CLAIMS:

1. A computer-implemented method (300) of analyzing and / or processing user generated fault report, comprising:obtaining (301), by a processing device (210), a fault report data comprising at least a first type of information including first data related to an identity of a lighting device (110), arranged in a target space (100), for which a fault has been reported in a fault report (13) created by a user;extracting (303), by the processing device, one or more identity markers (13a - 13c, 13f, 13g) for forming the identity of the lighting device through parsing the first data comprised in the obtained fault report data, the identity markers referring to those information or data which can be used to mark the identity of the lighting devices;determining (305), by the processing device, an initial identity of the lighting device based on the extracted one or more identity markers, the initial identity including a list of the extracted one or more identity markers of the faulty lighting device 110;obtaining (307), by the processing device from a lighting fixture database (230a), log data associated with a plurality of lighting fixtures, wherein the log data comprises recorded identity information of the plurality of lighting fixtures;comparing (309) the obtained log data to the determined initial identity of the lighting device in order to establish a match between the initial identity of the lighting device and a recorded identity information of a lighting fixture in the log data; and when a match was determined:confirming (311a) and / or updating (31 lb) the determined initial identity of the lighting device based on the determined match in order to establish (311) a unique identity of the lighting device.

2. The method (300) according to claim 1, wherein the fault report information further includes second data related to the reported fault associated with the identified lighting device, wherein the method further comprises:obtaining (313), by the processing device from a database (230b), a plurality of predetermined fault classes (CO - Cn) associated with lighting fixtures;2024PF8053929extracting (315) the reported fault for the identified lighting device through parsing the second data related to the reported fault and by comparing (316) the parsed second data to the obtained plurality of predetermined fault classes;identifying (317), between the plurality of the predetermined fault classes, a matching fault class indicative of a type of the extracted fault; andwhen the matching fault class is identified:classifying (319) the extracted fault in the identified fault class.

3. The method (300) according to any one of claims 1 - 2, wherein the method further comprises:providing (321a) a first input (401), by the processing device (210), to an artificial intelligence, Al, model (400) configured to determine outcome of diagnosis of reported faults associated with lighting fixtures, wherein the first input comprises the fault report data;utilizing the Al model by the processing device (210) for:extracting (303), the one or more identity markers (13a - 13c, 13f, 13g) and determining (305) the initial identity of the lighting device;wherein the method further comprises:obtaining (323a), by the processing device, a first output response (404a) from the Al model based on the provided first input, wherein the first output response comprises the initial identity of the lighting device, determined by the Al model.

4. The method (300) according to claim 1, wherein the method further comprises:providing (321b), by the processing device, a first input (401) to an Al model (400), wherein the first input comprises the obtained log data (403) and the initial identity of the lighting device (110);utilizing the Al model (400) by the processing device (210) for: comparing (309) the obtained log data to the determined initial identity of the lighting device and confirming (311a) and / or updating (31 lb) the determined initial identity of the lighting device;wherein the method further comprises:2024PF8053930obtaining (323b), by the processing device, an updated first output (404a”) of the Al model comprising the established unique identity of the lighting device, determined by the Al model.

5. The method (300) according to claim 2, wherein the method further comprises:providing (321c), by the processing device, the plurality of predetermined fault classes (CO - Cn) associated with lighting fixtures, and the established unique identity of the lighting device, as a second input (402) to an Al model (400);utilizing the Al model by the processing device (210) for:extracting (315a), the reported fault for the identified lighting device; identifying (317a), between the plurality of the predetermined fault classes, the matching fault class;classifying (319a), the extracted fault in the identified fault class; wherein the method further comprises:obtaining (323 c), by the processing device, a second output response (404b) from the Al model based on the provided second input, wherein the second output response comprises a summary of the extracted fault and the identified fault class of the extracted fault.

6. The method (300) according to any of claims 2 to 5, wherein the method further comprises:determining (325), by the processing device and / or by utilizing the Al model, outcome of diagnosis of the reported fault associated with the identified lighting device by:obtaining (327) a current operational status data of the identified lighting device from a server device in communication with the identified lighting device;comparing (329) the obtained current operational status data to the classified extracted fault; anddetermining (331) if the reported fault has been successfully resolved based on said comparison.

7. The method (300) according to any one of claims 3 - 6, wherein the Al model is a multimodal large model, MM-LM (400) configured to analyze and / or process information from multiple data modalities, and wherein the method further comprises:2024PF8053931providing (321d) a primary set of instructions to the MM-LM in order to contextually constrain the MM-LM to determine outcome of diagnosis of reported faults associated with lighting fixtures.

8. The method (300) according to any one of preceding claims, wherein the obtained fault report data further comprises a second type (13e, 13h) of fault report information, and wherein the first or the second type of fault report information include multiple data modalities being any one of textual data, image data or video data.

9. A computer program product comprising instructions which, when the program is executed by one or more processors of the processing device, causes the processing device to carry out the method according to any one of the claims 1 - 8.

10. A system (200) of analyzing and / or processing user generated fault report comprising a processing device (210) configured to:obtain a fault report data comprising at least a first type of information including first data related to an identity of a lighting device (110), arranged in a target space (100), for which a fault has been reported in a fault report created by a user;extract one or more identity markers (13a - 13c, 13f, 13g) for forming the identity of the lighting device through parsing the first data comprised in the obtained fault report data, the identity markers referring to those information or data which can be used to mark the identity of the lighting devices;determine an initial identity of the lighting device based on the extracted one or more identity markers, the initial identity including a list of the extracted one or more identity markers of the faulty lighting device 110;obtain, from a lighting fixture database (230a), log data associated with a plurality of lighting fixtures, wherein the log data comprises recorded identity information of the plurality of lighting fixtures;compare the obtained log data to the determined initial identity of the lighting device in order to establish a match between the initial identity of the lighting device and a recorded identity information of a lighting fixture in the log data; andwhen a match was determined:confirm and / or update the determined initial identity of the lighting device based on the determined match in order to establish a unique identity of the lighting device.2024PF805393211. The system (200) according to claim 10, wherein the fault report information further includes second data related to the reported fault associated with the identified lighting device (110), and wherein the processing device (210) is further configured to:obtain, from a database (230b), a plurality of predetermined fault classes (CO -Cn) associated with lighting fixtures;extract the reported fault for the identified lighting device through parsing the second data related to the reported fault and by comparing the parsed second data to the obtained plurality of predetermined fault classes;identify, between the plurality of the predetermined fault classes, a matching fault class indicative of a type of the extracted fault; andwhen a matching fault class is identified:classify the extracted fault in the identified fault class.

12. The system according to any one of claims 10 - 11, wherein the processing device (210) is further configured to:provide a first input (401) to an artificial intelligence, Al, model (400) configured to determine outcome of diagnosis of reported faults associated with lighting fixtures, the first input at least comprising the fault report data;utilize the Al model in order to extract, the one or more identity markers (13a - 13c, 13f, 13g) and determine the initial identity of the lighting device and / or compare the obtained log data to the determined initial identity of the lighting device and confirm and / or update the determined initial identity of the lighting device;wherein the processing device is further configured to:obtain a first output response (404a, 404a’, 404a”) from the Al model based on the provided first input, wherein the first output response comprises the initial identity and / or the unique identity of the lighting device determined by the Al model.

13. The system (200) according to claim 11, wherein the processing device (210) is further configured to:provide the plurality of predetermined fault classes (CO - Cn) associated with lighting fixtures, and the established unique identity of the lighting device, as a second input (402) to an Al model (400);utilize the Al model (400) in order to:2024PF8053933extract the reported fault for the identified lighting device;identify between the plurality of the predetermined fault classes, the matching fault class; andclassify the extracted fault in the identified fault class;wherein the processing device is further configured to:obtain a second output response (404b) from the Al model based on the provided second input, wherein the second output response comprises a summary of the extracted fault and the identified fault class of the extracted fault.

Images