Mobile application - Best practices
Depending on your mobile application, it is likely that many other features will be integrated alongside the image analysis provided by Deepomatic. However, there are several UX components that need to be taken into account in order to provide the best possible experience for field technicians:
  • The display of the status of an intervention - which checkpoints have already been validated or invalidated? for which checkpoints no photos have been taken by the technician?
  • The display of the results of the image analysis - trying to provide the user with as much information as possible (areas of the image, relative distance between various identified elements etc)

Status of an intervention

The visual quality control that is done on an intervention takes on its full meaning as the technician progresses through the intervention. It is therefore essential to be able to show him at any time the elements that have been validated, but especially those that have yet to be validated.
Above is a display proposal. As much as possible, it is useful to group the control points according to a certain business logic (chronological, geographical, etc.). However, these groupings are not configurable via the Deepomatic platform for the moment.

Results of image analysis

To best help the technician in his work, it is necessary to go further than sending him a simple validation or non-validation of a control point. It is possible to configure in a workflow the forwarding of areas of interest in the image to explain the result of the analysis for a given check point.
There is a display challenge when several control points are validated by the same photo. And there is an even bigger challenge when a control point needs several photos to be validated. In that case, it is necessary to introduce a new stage for a control point that we could called "in the process of validation".
In addition, it is very useful to let the user view all the photos that have been taken for a specific checkpoint at any time. One can imagine a kind of photo feed for an entire intervention or for a control point. For each of the photos, it can be interesting to display the result of the analysis.

Planning the future

One of the main challenges of integrating the Deepomatic solution into a mobile Field Service Management application is to foresee future evolutions of the workflow, and therefore of the API. The objective must be to avoid the need to make changes to the integration even if the list of control points evolves.
This is why we do not advise to have a too restrictive UX, which would guide the technician photo after photo.


You should also think carefully about the UX you wish to put in place to conciliate the objective of improving the quality of the interventions, but also the user experience of the technician, which should not be blocked if the image analysis fails or is wrong.
A good practice is to integrate a feedback button that allows the technician to make explicit the reason why he or she disregards the result of the image analysis. The feedback information is stored so that it is possible to exploit these images, and to prevent abuse by some technicians.
Last modified 1mo ago