Computer implemented method for improving safety at a jobsite

Abstract

The present invention relates to a computer implemented method for improving safety at a jobsite. The method comprising: upon receiving an indication that a fieldworker is about to join the jobsite, obtaining pre-job context data; determining, using a first machine learning model trained to output pre-job hazards predicted to be associated with a jobsite based on pre-job context data and having the obtained pre-job context data as input, one or more pre-job hazards predicted to be associated with the jobsite; and prior to that the fieldworker is arriving at the jobsite, displaying an indication of the one or more pre-job hazards at a display of a mobile computing device associated with the fieldworker.

Description

[0001] COMPUTER IMPLEMENTED METHOD FOR IMPROVING SAFETY AT A JOBSITE

[0002] Technical field

[0003] The present invention relates to improving safety at a jobsite. Especially a computer implemented method for improving safety at a jobsite is presented. The computer implemented method being suitable to be executed by a risk assessment system comprising one or more mobile computing devices and a remote server.

[0004] Background

[0005] In general, industries such as, but not limited to, construction industries, energy and utilities industries, or the like, involve day-to-day work (also be referred to as field work, field operations, or the like) to be performed at a location / work site. The work at the location has to be performed in compliance with field work management requirements defined based on state, local, and federal laws, a type of work, a work site, and so on. The field work management requirements include multiple regulations, codes and standards with regard to health and safety of workers, environmental protection, and quality management.

[0006] Performing risk assessment before, during and after a job may be encouraged or required in these types of industries. Video risk assessment can be used to identify risks in an area such as a jobsite, where work needs to be done efficiently without accidents or inconveniences slowing the work down or even stopping the work. Any field worker on the site can use a video risk assessment, VRA, application on a mobile phone or other mobile device that includes a camera, capturing video of the area in which they intend to work for video analysis.

[0007] Before arriving at the jobsite it would be beneficial for a fieldworker to have knowledge of potential hazards he or she may encounter at the jobsite. Having such knowledge may improve safety at the jobsite. This by, for example, letting the fieldworker have knowledge about special equipment to bring to the jobsite and / or have knowledge about actions to perform when arriving at the jobsite. It is however, hard to for the fieldworker in a reliable and trustworthy manner be notified about such potential hazards. Further, as the work progress at the jobsite conditions at the jobsite as well as site-specific hazards may change. Hence, there is a need for the fieldworker, and potentially also a manager supervising the jobsite, to be updated about hazards at the jobsite.

[0008] Summary of the invention

[0009] In view of the above, it is an object of the present invention to provide improved safety at a jobsite.

[0010] According to a first aspect computer implemented method for improving safety at a jobsite is provided. The method comprising: upon receiving an indication that a fieldworker is about to join the jobsite, obtaining pre-job context data; determining, using a first machine learning model trained to output pre-job hazards predicted to be associated with a jobsite based on pre-job context data and having the obtained prejob context data as input, one or more pre-job hazards predicted to be associated with the jobsite ; and prior to that the fieldworker is arriving at the jobsite, displaying an indication of the one or more pre-job hazards at a display of a mobile computing device associated with the fieldworker.

[0011] The method may comprise displaying an indication of the one or more pre-job hazards at a display of a computing device associated with a manager supervising the jobsite.

[0012] Receiving an indication that a fieldworker is about to join the jobsite may comprise receiving input from an I / O module of the mobile computing device indicating that the fieldworker is joining the jobsite.

[0013] Receiving an indication that a fieldworker is about to join the jobsite may comprise receiving a request from a manager of the jobsite or another fieldworker that the fieldworker is to join the jobsite.

[0014] An indication of a pre-job hazard may comprise details of the pre-job hazard.

[0015] An indication of a pre-job hazard may comprise an associated set of control measures for the pre-job hazard.

[0016] Upon a plurality of pre-job hazards predicted to be associated with the jobsite are outputted from the first machine learning model, the method may further comprise displaying an individual indication for each of the plurality of pre-job hazards at the display of the mobile computing device. The method may comprise requesting for the fieldworker to acknowledge contents of the indication of the one or more pre-job hazards.

[0017] Obtaining the pre-job context data may comprise receiving pre-job context data from one or more external application programming interfaces.

[0018] The pre-job context data may comprise data on historical jobs completed in the same / nearby location. In this context same / nearby means same street, same postcode or close proximity radius, e.g. 500m. Examples of such data on historical jobs are: job location, date of job start and completion, job type, job subtype, risk assessments completed on those jobs with validated hazards and controls.

[0019] The pre-job context data may comprise one or more of: data on nearby points of interest, weather data, traffic data, roadworks data, geographic information system, GIS, data, and telematics data.

[0020] The pre-job context data may comprise historical personal behavior data for the fieldworker.

[0021] The method may comprise: upon, the fieldworker is arriving at the jobsite, enabling the fieldworker to capture an audio-visual presentation of a site of the job using a video module of the mobile computing device; determining, using a second machine learning model trained to output site specific hazards based on video and / or audio data and having the audio-visual presentation of the jobsite as input, one or more site-specific hazards; and displaying an indication of the one or more site-specific hazards at the display of the mobile computing device.

[0022] The second machine learning model may additionally trained based on sitespecific context data, wherein the method further comprises in addition to inputting the audio-visual presentation of the jobsite to the second machine learning model also inputting obtained site-specific context data to the second machine learning model.

[0023] An indication of a site-specific hazard may comprise details of the site-specific hazard.

[0024] An indication of a site-specific hazard may comprise an associated set of control measures for the site-specific hazard.

[0025] Upon a plurality of site-specific hazards are outputted from the second machine learning model, the method may comprise displaying an individual indication for each of the plurality of site-specific hazards at the display of the mobile computing device. The method may comprise requesting for the fieldworker to acknowledge contents of the indication of the one or more site-specific hazards.

[0026] Obtaining the site-specific context data may comprise receiving context data from one or more external application programming interfaces.

[0027] The site-specific context data may comprise one or more of: data on nearby points of interest, weather data, traffic data, roadworks data, geographic information system, GIS, data, and telematics data.

[0028] The site-specific context data may comprise historical personal behavior data for the fieldworker.

[0029] The method may comprise: monitoring whether there has been a change in the site-specific context data; upon there has been a change in the site-specific context data, obtaining updated site-specific context data; inputting the obtained updated sitespecific context data to the second machine learning model, thereby generating an updated site-specific hazard prediction; and displaying an indication of one or more updated site-specific hazards at the display of the mobile computing device.

[0030] The method may comprise prompting the fieldworker to capture a new audiovisual presentation of the jobsite using the video module of the mobile computing device; wherein generating the updated site-specific hazard prediction further comprises inputting the new audio-visual presentation to the second machine learning model.

[0031] The method may comprise periodically prompting the fieldworker to capture a new audio-visual presentation of the jobsite using the video module of the mobile computing device; generating an updated site-specific hazard prediction by inputting the new audio-visual presentation to the second machine learning model; and displaying an indication of one or more updated site-specific hazards at the display of the mobile computing device.

[0032] Periodically prompting may be performed on an daily basis, such as a 2-5 times during a working day.

[0033] According to a second aspect a non-transitory computer-readable storage medium is provided. The non-transitory computer-readable storage medium having stored thereon instructions for implementing the method according to the first aspect, when executed on a device having processing capabilities The above mentioned features of the method, when applicable, apply to this second aspect as well. In order to avoid undue repetition, reference is made to the above.

[0034] A further scope of applicability will become apparent from the detailed description given below. However, it should be understood that the detailed description and specific examples are given by way of illustration only.

[0035] It is to be understood that the terminology used herein is for purpose of describing particular embodiments only, and is not intended to be limiting. It must be noted that, as used in the specification and the appended claim, the articles "a," "an," "the," and "said" are intended to mean that there are one or more of the elements unless the context clearly dictates otherwise. Thus, for example, reference to "a unit" or "the unit" may include several devices, and the like. Furthermore, the words "comprising", “including”, “containing” and similar wordings does not exclude other elements or steps.

[0036] Brief description of the drawings

[0037] The above and other aspects will now be described in more detail, with reference to appended figures. The figures should not be considered limiting; instead they are used for explaining and understanding.

[0038] As illustrated in the figures, the sizes of layers and regions may be exaggerated for illustrative purposes and, thus, are provided to illustrate the general structures. Like reference numerals refer to like elements throughout.

[0039] Fig. 1 illustrate a risk assessment system and a fieldworker interacting with a mobile computing device of the risk assessment system at a jobsite comprising a plurality of hazards.

[0040] Fig. 2 is a block diagram of a method for providing potential hazards a fieldworker may encounter at a jobsite to the fieldworker before he or she is arriving at the jobsite.

[0041] Figs 3a-3d illustrate screenshots of different implementations of different steps in the method of Fig. 2.

[0042] Fig. 4 is a block diagram of a method for onsite video risk assessment.

[0043] Fig. 5 illustrates an example of displaying the one or more site-specific hazards determined with the method of Fig. 4 to the fieldworker. Fig. 6 is a block diagram of a method for post video risk assessment updating of hazards.

[0044] Detailed description

[0045] The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which currently preferred embodiments of the invention are shown. This invention may, however, be embodied in many different forms.

[0046] In connection with Fig. general concepts of risk assessment using a mobile computing device 102 associated with a field worker 100 will be discussed. The field worker 100 may capture an audio-visual presentation of a jobsite using the mobile computing device 102. The audio-visual presentation comprising both video and audio data. The video data comprises visual information in a series of frames. The audio data comprise audio information. Risks may be assessed based on the video data, the audio data or both. A video to be used for video risk assessment, VRA, is typically the order of minutes in duration. Especially, the duration of the video for VRA shall be sufficiently long for the field worker 100 to indicate the hazards that they can see and any other notable things that they can see or know about the jobsite.

[0047] Fig. 1 shows an example field worker 100, who will capture a video at a jobsite 120 in order for a video risk assessment to be performed. Figure 1 depicts an example mobile computing device 102 comprising a video module 104. The video module 104 comprises a video recorder configured to record video. In this example, a lens of the video recorder is shown on the back of the mobile computing device 102. Fig. 1 also shows the front of the mobile computing device 102 which in this example comprises a display 106. The mobile computing device 102 may additionally or alternatively include a lens of the video recorder on the front side - for example, to allow a field worker 100 to capture video including their face using a front-facing camera. Further, the mobile computing device 102 typically also comprise a microphone 110 configured to record audio. In order for the video module 104 to capture both video data and audio data it typically comprises both the video recorder (having either a back-facing camera, a front-facing camera or both) and the microphone 110. The mobile computing device 102 may further comprise a speaker 110a configured to play audio. The mobile computing device 102 comprises a processor 108 and memory 112. The memory 112 comprises an operating system configured to operate the mobile computing device 102. The memory 112 may store one or more mobile applications.

[0048] The mobile computing device 102 may be a mobile phone, a tablet, a laptop or another suitable mobile computing device. A computing device that the field worker 100 cannot carry around at a jobsite 120 would not be suitable for the present purpose. A computing device moveable on wheels or the like may be used but may not be the most elegant solution to capturing video of a jobsite versus a handheld mobile computing device. The mobile computing device 102 may be a digital camera equipped with a processor 108 and memory 112 such that a mobile application can be run on the digital camera.

[0049] The video module 104, typically also including a microphone 110, and the speaker 110a may be operatively connected to the memory 112 and the processor 108. The mobile computing device 102 further comprise controlling circuitry 114 configured to control operations of the mobile computing device 102. The controlling circuitry 114 is further configured to execute functions to be performed by the mobile computing device 102. One or more such functions may be in the form of steps of a method presented in this disclosure. The mobile device 102 comprise a transmitter and receiver or transceiver 116 configured to send and receive digital signals. The display 106 may comprise or be provided additionally to an I / O module 106a, which may comprise a keyboard or touchscreen to allow a user to input information into the mobile computing device 102. The I / O module 106a may comprise the display 106 and the speaker 110a, for example, such that information can be presented to the user as an output of analysis. The information may take the form of text, graphical representations such as charts or diagrams, or audio - for example a spoken message in a relevant language to the user or sounds such as bells or alarms or the like. The microphone 110 may be part of the I / O and may allow to field worker 100 to control operations of the mobile computing device 102 using voice commands, for example.

[0050] In Fig. 1, the jobsite 120 is shown comprising three example hazards 122: cloudy weather (which may indicate that rain will fall at the jobsite or may reduce visibility), an area where entry is prohibited, a road and an electrical hazard, for example power lines. These are merely examples of hazards that could occur at a jobsite 120 that would affect how a job is carried out. The field worker 100 may capture an audio-visual presentation of the jobsite 120 including one or more hazards. For video risk assessment purposes, the field worker 100 may spend time showing a hazard in detail (from a safe distance or in a safe manner) or may narrate their audiovisual presentation to indicate that they have perceived the hazard - for example “I am showing a road that passes close to the jobsite. I have observed that the road is currently busy”.

[0051] The mobile computing device 102 is typically part of a risk assessment system 300 also comprising a remote server 302. The remote server 302 may be a single server device. Alternatively, the remote server 302 may be a cloud-based server. The remote server 302 comprises circuitry 306, a processor 308 and a memory 312. The circuitry 306 is configured to execute different functions to be executed by the remote server 302 . One or more of such functions may be in the form of steps of a method presented in this disclosure. Further, functions forming steps of a method presented in this disclosure may be executed partly by the controlling circuitry 114 of the mobile computing device 102 and partly by the circuitry 306 of the server 302. Results from a function executed by the controlling circuitry 114 of the mobile computing device 102 may be used by a function executed by the circuitry 306 of the server 302 and vice versa. For example, functions for providing an output to the fieldworker 100 is typically executed by the controlling circuitry 114 of the mobile computing device 102. The mobile computing device 102 is connectable to the remote server 302 using a communication network. The communication network is typically a wireless network such as a cellular network, or a Wi-Fi network.

[0052] Before arriving at the jobsite it would be beneficial for the fieldworker 100 to have knowledge of potential hazards he or she may encounter at the jobsite 120. Having such knowledge may improve safety at the jobsite 120. This by, for example, letting the fieldworker 100 have knowledge about special equipment to bring to the jobsite 120 and / or have knowledge about actions to perform when arriving at the jobsite 120. It is however, hard to for the fieldworker 100 in a reliable and trustworthy manner be notified about such potential hazards. In the following such potential hazards will be referred to as pre-job hazards predicted to be associated with the jobsite 120. In connection with this a computer implemented method for the fieldworker 100 to be notified of pre-job hazards predicted to be associated with the jobsite 120 will now be presented. In connection with Fig. 2 a computer implemented method 200 for providing potential hazards a fieldworker 100 may encounter at the jobsite 120 to the fieldworker 100 before he or she is arriving at the jobsite 120 will be discussed. Below such potential hazards will be referred to pre-job hazards predicted to be associated with the jobsite 120. Fig. 2 is a flow chart illustrating the steps of the method 200. Below, the different steps is described in more detail. Even though illustrated in a specific order, the steps of the method 200 may be performed in any suitable order, in parallel, as well as multiple times. Some of the steps, or even all steps, of the method 200 may be executed by a mobile computing device, typically the mobile computing device 102 associated with the fieldworker 100. Some of the steps may be executed by the remote server 302 and / or a device 320 associated with a manager responsible for the jobsite 120.

[0053] Upon receiving S201 an indication that a fieldworker 100 is about to join the jobsite 120, pre-job context data is obtained S202. Pre-job context data is typically received from one or more internal and external application programming interfaces. For example, historical data on jobs and user behaviour are considered internal as it is proprietary data of the client. Some examples of external application programming interfaces are a weather service, a traffic service, geographic information system, GIS, service, ongoing job service and telematics data service. Pre-job context data can be any type of data that might indicate a presence of a potential hazard at a specific site, e.g. the jobsite 120 the fieldworker 100 is about to join.

[0054] The pre-job context data may comprise data on historical jobs completed in the same / nearby location (i.e. same street, same postcode or close proximity radius, e.g. 500m). Examples of such data on historical jobs are: job location, date of job start and completion, job type, job subtype, and risk assessments completed on those jobs with validated hazards and controls.

[0055] The pre-job context data may comprise one or more of: data on nearby points of interest, for example bus stops, schools, hospitals, shopping centers, etc.; traffic data, for example intensity, speed, incidents, etc.; roadworks data, for example traffic restrictions, road closures, etc.; weather data, for example rain, snow, wind, ice, heat, storms, etc.; GIS data, for example data related to underground assets, gas main location, ground condition, etc.; telematics data, for example historical data on how vehicles has been driven on the day or before.

[0056] The pre-job context data may alternatively or in addition comprise historical personal behavior data for the fieldworker 100. Examples of such historical personal behavior data for the fieldworker 100 are: experience level, recent near misses or incidents, number of hours worked during the day, number of hours worked in the last 7 days, quality of risk assessments submitted before, and genuine interactions with hazards and controls before.

[0057] In more detail, some aspects of the historical personal behavior data may update slowly, such as “experience”, but other aspects will update as the day goes, such as “number of hours worked” and might start making impact on risk assessment as the fieldworker is becoming more and more fatigued. Hence, the historical personal behavior data may comprise both relatively static data, such as experience level, and dynamic data, such as number of hours worked during the day. Accordingly, in case of the historical personal behavior data comprising dynamic data such dynamic data may also be used in the risk assessment.

[0058] Receiving S201 an indication that the fieldworker 100 is about to join the jobsite 120 may comprise receiving input from the I / O module 106a of the mobile computing device 102 indicating that the fieldworker 100 is joining the jobsite 120. Receiving such input may be triggered by prompting, e.g. on the display 106 of the mobile computing device 102, the fieldworker 100 to input an indication that she or he is joining the jobsite 120. Additionally, or in combination, receiving S201 an indication that the fieldworker 100 is about to join the jobsite 120 may comprise that a manager of the jobsite 120 or another fieldworker is requesting the fieldworker 100 to join the jobsite 120. Such a request is then typically made using a device 320 associated with the manager or a device associated with the another fieldworker. Such a device 320 associated with the manager or such a device associated with the another fieldworker is then also communicably connected to the risk assessment system 300.

[0059] In connection with Figs 3a and 3b an example of receiving S201 an indication that the fieldworker 100 is about to join the jobsite 120 is illustrated. As illustrated in Fig. 3a the fieldworker 100 may be provided with potential jobsites to join. The fieldworker 100 is then selecting one of the jobsites 120 to join, this is illustrated in Fig. 3b.

[0060] Next, the obtained pre-job context data is inputted into a first machine learning model. The first machine learning model is trained to output pre-job hazards predicted to be associated with a jobsite based on pre-job context data inputted thereto. The first machine learning model is trained based on historical risk assessments, preferably historical risk assessments done in an area nearby the jobsite 120. In this context nearby means one or more of same street, same postcode, within a radius of 500m. The first machine learning model may be executed on the mobile computing device 102 associated with the fieldworker 100. Alternatively, or in combination, the first machine learning model may be executed on the remote server 302. The output from the first machine learning model is determined S204 to be one or more pre-job hazards predicted to be associated with the jobsite 120.

[0061] An indication of the one or more pre-job hazards is displayed S206 at the display 310 of the mobile computing device 102 associated with the fieldworker 100. An example of displaying S206 the one or more pre-job hazards to the fieldworker 100 is illustrated in Fig. 3c. The displaying is made prior to that the fieldworker 100 is arriving at the jobsite 120. Additionally, an indication of the one or more pre-job hazards may as well be displayed at the device 320 associated with a manager supervising that jobsite 120. The indication of the one or more pre-job hazards comprises details of the respective pre-job hazard. Exampled of such details are illustrated in Fig. 3c. The indication of the one or more pre-job hazards may further comprise an associated set of control measures for each of the pre-job hazards. Such a set of control measures may comprise one or more control measures. As illustrated in connection with Fig. 3d, some examples of details of a pre-job hazard and an associated set of control measures are:

[0062] • School 0.2km from site. Be cautious of school zones and children crossing. Ensure barriers in place.

[0063] • Repair job 0.8km away; Utility work nearby; coordinate to avoid issues.

[0064] • Cold weather. Wear appropriate clothing. Ensure equipment is safe.

[0065] The control measures may be accessed by the fieldworker requesting such control measures to be seen, e.g. by clicking the “View site conditions” button in Fig. 3c. In case there is more than one pre-job hazard predicted to be associated with the jobsite 120 outputted from the first machine learning model, each such pre-job hazard may be displayed S206 as an individual indication. As discussed above, such displaying S206 is made at the display 106 of the mobile computing device 102 associated with the fieldworker 100. One or more of the individual indications of the pre-job hazards may as well be displayed to the manager of the jobsite 120 on the device 320 associated with the manager.

[0066] In connection with displaying S206 the indications of the one or more pre-job hazards a request for the fieldworker 100 to acknowledge the contents of the respective indication may be made. Such acknowledgement may be prompted for by using a virtual close / next button on the display 106.

[0067] The method 200 is typically implemented as a computer program to be executed in the risk assessment system 300. Some or all of the steps of the method 200 may be executed at the mobile computing device 102 associated with the fieldworker 100. Some of the steps of the method 200 may be executed at the remote server 302. Some of the steps of the method 200 may be executed at the device 320 associated with the manager. The computer program for implementing the method 200 may be stored on a non-transitory computer-readable storage medium.

[0068] Once arriving at the jobsite 120, onsite risk assessment would be beneficial to be performed. One type of risk assessment is video risk assessment, VRA. In connection with Fig 4 a method 400 for onsite VRA will be discussed. The method 400 for onsite VRA can be executed after the method 200 for providing one or more pre-job hazards discussed above in connection with Fig. 2. Fig. 4 is a flow chart illustrating the steps of the method 400. Below, the different steps is described in more detail. Even though illustrated in a specific order, the steps of the method 400 may be performed in any suitable order, in parallel, as well as multiple times. Some of the steps, or even all steps, of the method 400 may be executed by a mobile computing device, typically the mobile computing device 102 associated with the fieldworker 100. Some of the steps of the method 400 may be executed by the remote server 302 and / or the device 320 associated with the manager.

[0069] Upon an indication that the fieldworker 100 has arrived S401 at the jobsite 120, the fieldworker 100 will be enabled to capture S402 an audio-visual presentation of the jobsite 120. This by using the video module 104 of the mobile computing device 102. As discussed above, the audio-visual presentation comprising both video data and audio data. The indication that the fieldworker 100 has arrived S401 at the jobsite 120 may be given by comparing a location of the jobsite 120 with a location of the mobile computing device 102 associated with the fieldworker 100. If the location of the jobsite 120 corresponds with the location of the mobile computing device 102 it is considered that the fieldworker 100 is at the jobsite. In this context a location may be seen as a GPS location. Enabling S402 the fieldworker 100 to capture an audio-visual presentation of the jobsite 120 may comprise prompting the fieldworker 100 to actually perform the capturing of the audio-visual presentation. Such prompting may be made by alerting the fieldworker 100 to actually perform the capturing of the audio-visual presentation. During the capturing of the audio-visual presentation the fieldworker 100 may be instructed about situations and / or parts of the jobsite 120 to be contained in the audio-visual presentation. What situations and / or parts of the jobsite 120 to be contained in the audio-visual presentation may be selected based on the pre-job hazards determined in the method 200 for providing one or more pre-job hazards.

[0070] Next, the one or more site-specific hazards is determined S404. This by inputting the captured audio-visual presentation of the jobsite 120 to a second machine learning model. The second machine learning model is trained to output site-specific hazards based on video data and / or audio data. The second machine learning model is trained based on historical risk assessments. The second machine learning model is working with input in form of risk assessment video and audio in the form of an audiovisual presentation. The output from the second machine learning model is a list of sitespecific hazards. The second machine learning model may be executed on a mobile computing device, such as the mobile computing device 102 associated with the fieldworker 100. Alternatively, or in combination, the second machine learning model may be executed on the remote server 302.

[0071] The second machine learning model may additionally be trained based on sitespecific context data. If so, the method 400 may further comprise, in addition to inputting the audio-visual presentation of the jobsite 120 to the second machine learning model also inputting obtained S403 site-specific context data into the second machine learning model. The output from the second machine learning model still being one or more site specific hazards. The site-specific context data may be obtained S403 upon the field worker is arriving S401 at the job. The site-specific context data may be received from one or more external application programming interfaces. Some examples of such external application programming interfaces are a weather service, a traffic service, geographic information system, GIS, service, ongoing job service and telematics data service. Site-specific context data can be any type of data that might indicate a presence of a potential hazard at a specific site, e.g. the jobsite 120. The site-specific context data may comprises one or more of: data on nearby points of interest, for example bus stops, schools, hospitals, shopping centers, etc.; traffic data, for example intensity, speed, incidents, etc.; roadworks data, for example traffic restrictions, road closures, etc.; weather data, for example rain, snow, wind, ice, heat, storms, etc.; GIS data, for example data related to underground assets, gas main location, ground condition, etc.; telematics data, for example historical data on how vehicles has been driven on the day or before.

[0072] The site-specific context data may alternatively or in addition comprise historical personal behavior data for the fieldworker 100. Examples of such historical personal behavior data for the fieldworker 100 are: experience level, recent near misses or incidents, number of hours worked during the day, number of hours worked in the last 7 days, quality of risk assessments submitted before, and genuine interactions with hazards and controls before.

[0073] The fieldworker 100 can then review and confirm the site-specific hazards, add more hazards or remove any irrelevant ones. This since an indication of the one or more site-specific hazards is displayed S406 at the display 106 of the mobile computing device 102. An example of displaying S406 the one or more site-specific hazards to the fieldworker 100 is illustrated in Figs 5a and 5b. Additionally, an indication of the one or more site-specific hazards may as well be displayed at the device 320 associated with a manager supervising that jobsite 120. The indication of the one or more site-specific hazards comprises details of the respective site-specific hazard. Examples of such details are illustrated in Fig. 5a. The indication of the one or more site-specific hazards may further comprise an associated set of control measures for each of the site-specific hazards. Such a set of control measures may comprise one or more control measures. Examples of such set of control measures are illustrated in Fig. 5b. Other examples of details of a site-specific hazard and an associated set of control measures are:

[0074] • Trees and hedgerows. Ensure works has minimal effect on surrounding trees and hedges (refer to environmental handbook). No cutting roots larger than 25mm.

[0075] • Moderate Volume of Cars or Light Commercial Vehicles. Safeguard adequate safety zones around works. Consider Traffic Management as per New Roads and Street Works Act, NRSWA.

[0076] • Pedestrian and vehicles. Safeguard appropriate barriers / signage in place as per NRSWA. Consider additional walkway for pedestrians. Consider curb ramps. Maintain access to properties. Secure all tools and ladders / clear signage in place / secure barriers in place. Segregation and safety zones in place, clearly signed as per NRSWA.

[0077] • Restricted visibility by work vehicles or local environment. Consider traffic management arrangements as per NRSWA.

[0078] • High Winds. Appropriate weights for barriers / signs must be used. Beware of falling debris. Remove branded netting from Heras fencing.

[0079] • High Crime Area. Be aware of potential vandalism. Ensure vehicles kept locked when working. Keep all tools and equipment secure.

[0080] • Pin Bar in use. Accurate up to date utility drawings available. The horizontal searcher bar must only be used within the confines of an excavation when the gas pipe has been located. Use CAT and Genny in all modes and mark up P,G,A & R.

[0081] • Adverse Weather. Drink warm fluids. Keep Hydrated. Regular breaks. Sunscreen on exposed skin. Warm up prior to manual work. Wear layered clothing.

[0082] • Gas Escape or Gas in Atmosphere. Monitor as per procedures. No smoking. Remove sources of ignition. Wear working PAM.

[0083] The control measures may e.g. be accessed by the fieldworker 100 by interacting with the respective detailed hazard in the screen illustrated in Fig. 5.

[0084] In case there is more than one site-specific hazard outputted from the second machine learning model, each such site-specific hazard may be displayed S406 as an individual indication. As discussed above, such displaying S406 is made at the display 106 of the mobile computing device 102 associated with the fieldworker 100. One or more of the individual indications of the site-specific hazards may as well be displayed to the manager of the jobsite 120 on the device 320 associated with the manager.

[0085] In connection with displaying S406 the indications of the one or more sitespecific hazards a request for the fieldworker 100 to acknowledge the contents of the respective indication may be made. Such acknowledgement may be prompted for by using a virtual close / next button on the display 106.

[0086] The method 400 is typically implemented as a computer program to be executed in the risk assessment system 300. Some or all of the steps of the method 400 may be executed at the mobile computing device 102 associated with the fieldworker 100. Some of the steps of the method 400 may be executed at the remote server 302. Some of the steps of the method 400 may be executed at the device 320 associated with the manager. The computer program for implementing the method 400 may be stored on a non-transitory computer-readable storage medium.

[0087] As the work progress at the jobsite 120 conditions at the jobsite 120 as well as site-specific hazards may change. Hence, there is a need for the fieldworker 100, and potentially also the manager supervising the jobsite 120, to be updated about new sitespecific hazards and also about which site-specific hazards that may still remain. Possibly also which site-specific hazards that no longer are present shall be presented to the fieldworker 100 and / or the manager supervising the jobsite 120.

[0088] In connection with Fig. 6 a method 600 for post VRA updating of hazards will be discussed. The method 600 for post VRA updating of hazards can be executed after the method 400 for onsite VRA discussed above in connection with Fig. 4. Fig. 6 is a flow chart illustrating the steps of the method 600. Below, the different steps is described in more detail. Even though illustrated in a specific order, the steps of the method 600 may be performed in any suitable order, in parallel, as well as multiple times. Some of the steps, or even all steps, of the method 600 may be executed by a mobile computing device, typically the mobile computing device 102 associated with the fieldworker 100. Some of the steps of the method 400 may be executed by the remote server 302 and / or the device 320 associated with the manager.

[0089] Monitoring S602 whether there has been a change in the site-specific context data is performed. The monitoring S602 may be made periodically, e.g. once a hour. The monitoring S602 may be made by making a new call to the one or more external application programming interfaces for updated site-specific context data. As discussed above, some examples of such external application programming interfaces are a weather service, a traffic service, geographic information system, GIS, service, ongoing job service and telematics data service. Site-specific context data can be any type of data that might indicate a presence of a potential hazard. The site-specific context data may comprises one or more of: data on nearby points of interest, for example bus stops, schools, hospitals, shopping centers, etc.; traffic data, for example intensity, speed, incidents, etc.; roadworks data, for example traffic restrictions, road closures, etc.; weather data, for example rain, snow, wind, ice, heat, storms, etc.;

[0090] GIS data, for example data related to underground assets, gas main location, ground condition, etc.; telematics data, for example historical data on how vehicles has been driven on the day or before.

[0091] The site-specific context data may alternatively or in addition comprise historical personal behavior data for the fieldworker 100. Examples of such historical personal behavior data for the fieldworker 100 are: experience level, recent near misses or incidents, number of hours worked during the day, number of hours worked in the last 7 days, quality of risk assessments submitted before, and genuine interactions with hazards and controls before.

[0092] Upon there has been a change in the site-specific context data, obtaining S603 updated site-specific context data. Based on the updated site-specific context data an updated site-specific hazard prediction is generating S606 by the second machine learning model using the obtained updated site-specific context data as input.

[0093] In addition, or alternatively, to using the updated site-specific context data as input to the second machine learning model, the fieldworker 100 may be prompted S604 to capture a new audio-visual presentation of the jobsite 120 to be used for generating S606 the updated site-specific hazard prediction. Hence, the new audiovisual presentation of the jobsite 120 captured by the fieldworker 100 may also be used as input to the second machine learning model for generating S606 the updated sitespecific hazard prediction. Hence, generating S606 the updated site-specific hazard prediction may be performed by jointly inputting both the new audio-visual presentation and the updated site-specific context data to the second machine learning model. Alternatively, the updated site-specific hazard prediction may be generated S606 by inputting either only the new audio-visual presentation or only the updated site-specific context data to the second machine learning model.

[0094] The manager supervising the jobsite may also be notified about a change in site-specific context data. Hence, the manager may, via the device 320 associated with the manager, control the prompting S604 of the fieldworker 100 to capture a new audio-visual presentation of the jobsite 120 to be used for generating S606 the updated site-specific hazard prediction.

[0095] Upon the updated site-specific hazard prediction has been generated S606, an indication of one or more updated site-specific hazards is displayed S608 at the display 106 of the mobile computing device 102. In addition to updated site-specific hazards also remaining site-specific hazards may be displayed. Further, no longer existing sitespecific hazards may be displayed as no longer existing site-specific hazards.

[0096] Hence, the fieldworker 100 can review and confirm new and remaining sitespecific hazards. This since an indication of the one or more site-specific hazards is displayed S608 at the display 106 of the mobile computing device 102. Additionally, an indication of the one or more new and remaining site-specific hazards may as well be displayed at the device 320 associated with the manager supervising the jobsite 120. As discussed above, the indication of the one or more new and remaining site-specific hazards comprises details of the respective site-specific hazard. As also discussed above, the indication of the one or more site-specific hazards may further comprise an associated set of control measures for each of the site-specific hazards. Such a set of control measures may comprise one or more control measures. Reference is made to the above discussions for examples of details and control measures for potential sitespecific hazards.

[0097] In case there is more than one new or remaining site-specific hazard outputted from the second machine learning model, each such site-specific hazard may be displayed S608 as an individual indication. As discussed above, such displaying S608 is made at the display 106 of the mobile computing device 102 associated with the fieldworker 100. One or more of the individual indications of the site-specific hazards may as well be displayed to the manager of the jobsite 120 on the device 320 associated with the manager.

[0098] In connection with displaying S608 the indications of the one or more sitespecific hazards a request for the fieldworker 100 to acknowledge the contents of the respective indication may be made. Such acknowledgement may be prompted for by using a virtual close / next button on the display 106.

[0099] The method 600 is typically implemented as a computer program to be executed in the risk assessment system 300. Some or all of the steps of the method 600 may be executed at the mobile computing device 102 associated with the fieldworker 100. Some of the steps of the method 600 may be executed at the remote server 302. Some of the steps of the method 600 may be executed at the device 320 associated with the manager. The computer program for implementing the method 600 may be stored on a non-transitory computer-readable storage medium.

[0100] The person skilled in the art realizes that the present invention by no means is limited to what is explicitly described above. On the contrary, many modifications and variations are possible within the scope of the appended claims.

[0101] Additionally, variations can be understood and effected by the skilled person in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

Claims

CLAIMS1. A computer implemented method for improving safety at a jobsite (120), the method comprising: upon receiving (S201) an indication that a fieldworker (100) is about to join the jobsite (120), obtaining (S202) pre-job context data; determining (S204), using a first machine learning model trained to output prejob hazards predicted to be associated with a jobsite based on pre-job context data and having the obtained pre-job context data as input, one or more pre-job hazards predicted to be associated with the jobsite (120); and prior to that the fieldworker (100) is arriving at the jobsite (120), displaying (S206) an indication of the one or more pre-job hazards at a display (310) of a mobile computing device (102) associated with the fieldworker (100).

2. The method according to claim 1, further comprising displaying an indication of the one or more pre-job hazards at a display (322) of a computing device (320) associated with a manager supervising the jobsite (120).

3. The method according to claim 1 or 2, wherein receiving (S201) an indication that a fieldworker (100) is about to join the jobsite (120) comprises receiving input from an I / O module (106a) of the mobile computing device (102) indicating that the fieldworker (100) is joining the jobsite (120).

4. The method according to any one of claims 1-3, wherein receiving (S201) an indication that a fieldworker (100) is about to join the jobsite (120) comprises receiving a request from a manager of the jobsite (120) or another fieldworker that the fieldworker (100) is to join the jobsite (120).

5. The method according to any one of claims 1-4, wherein an indication of a pre-job hazard comprises details of the pre-job hazard.

6. The method according to claim 5, wherein an indication of a pre-job hazard further comprises an associated set of control measures for the pre-job hazard.

7. The method according to any one of claims 1-6, upon a plurality of pre-job hazards predicted to be associated with the jobsite (120) are outputted from the first machine learning model, the method further comprises displaying an individual indication for each of the plurality of pre-job hazards at the display (106) of the mobile computing device (102).

8. The method according to any one of claims 1-7, further comprising requesting for the fieldworker (100) to acknowledge contents of the indication of the one or more pre-job hazards.

9. The method according to any one of claims 1-8, wherein obtaining the pre-job context data comprises receiving pre-job context data from one or more external application programming interfaces.

10. The method according to any one of claims 1-9, wherein the pre-job context data comprises one or more of: data on nearby points of interest, weather data, traffic data, roadworks data, geographic information system, GIS, data, and telematics data.

11. The method according to any one of claims 1-10, wherein the pre-job context data comprises historical personal behavior data for the fieldworker (100).

12. The method according to any one of claims 1-11 wherein pre-job context data comprises data on historical jobs completed in the same / nearby location.

13. The method according to any one of claims 1-12, further comprising: upon, the fieldworker (100) is arriving (S401) at the jobsite (120), enabling the fieldworker to capture (S402) an audio-visual presentation of a site of the job using a video module (104) of the mobile computing device (102); determining (S404), using a second machine learning model trained to output site specific hazards based on video and / or audio data and having the audio-visual presentation of the jobsite (120) as input, one or more site-specific hazards; and displaying (S406) an indication of the one or more site-specific hazards at the display (106) of the mobile computing device (102).

14. The method according to claim 13, wherein the second machine learning model is additionally trained based on site-specific context data, wherein the method further comprises: in addition to inputting the audio-visual presentation of the jobsite (120) to the second machine learning model also inputting obtained (S403) site-specific context data to the second machine learning model.

15. The method according to claim 13 or 14, wherein an indication of a site-specific hazard comprises details of the site-specific hazard.

16. The method according to claim 15, wherein an indication of a site-specific hazard further comprises an associated set of control measures for the site-specific hazard.

17. The method according to any one of claims 13-16, upon a plurality of site-specific hazards are outputted from the second machine learning model, the method further comprises displaying an individual indication for each of the plurality of site-specific hazards at the display (106) of the mobile computing device (102).

18. The method according to any one of claims 13-17, further comprising requesting for the fieldworker (100) to acknowledge contents of the indication of the one or more sitespecific hazards.

19. The method according to any one of claims 14-18, wherein obtaining the sitespecific context data comprises receiving context data from one or more external application programming interfaces.

20. The method according to any one of claims 14-19, wherein the site-specific context data comprises one or more of: data on nearby points of interest, weather data, traffic data, roadworks data, geographic information system, GIS, data, and telematics data.

21. The method according to any one of claims 14-20, wherein the site-specific context data further comprises historical personal behavior data for the fieldworker (100).

22. The method according to any one of claims 14-21 , further comprising:monitoring (S602) whether there has been a change in the site-specific context data; upon there has been a change in the site-specific context data, obtaining (S603) updated site-specific context data; inputting the obtained updated site-specific context data to the second machine learning model, thereby generating (S606) an updated site-specific hazard prediction; and displaying (S608) an indication of one or more updated site-specific hazards at the display (106) of the mobile computing device (102).

23. The method according to claim 22, further comprising prompting (S604) the fieldworker (100) to capture a new audio-visual presentation of the jobsite (120) using the video module (104) of the mobile computing device (102); wherein generating (S606) the updated site-specific hazard prediction further comprises inputting the new audio-visual presentation to the second machine learning model.

24. The method according to any one of claims 13-21 , further comprising: periodically prompting (S604) the fieldworker (100) to capture a new audiovisual presentation of the jobsite (120) using the video module (104) of the mobile computing device (102); generating (S606) an updated site-specific hazard prediction by inputting the new audio-visual presentation to the second machine learning model; and displaying (S608) an indication of one or more updated site-specific hazards at the display (106) of the mobile computing device (102).

25. The method according to claim 24, wherein the periodically prompting (S604) is performed on an daily basis, such as a 2-5 times during a working day.

26. A non-transitory computer-readable storage medium having stored thereon instructions for implementing the method according to any one of claims 1-25, when executed on a device having processing capabilities.

Images