72 lines
4.7 KiB
Typst
72 lines
4.7 KiB
Typst
#import "/metadata.typ": *
|
|
#import "objectives.typ":*
|
|
|
|
#pagebreak()
|
|
= #i18n("conclusion-title", lang:option.lang) <sec:conclusion>
|
|
|
|
== Project conclusion
|
|
Considering the objectives in @sec:objectives, the project takeaways are as follows.
|
|
|
|
The solution provided at the project's end allows to store the temperature, humidity and @co2 level in the equipped room.
|
|
Hence, the objective MO1 is fulfilled.
|
|
|
|
Independently, a user interface is made available to display the room environmental data.
|
|
As this user interface request the database, it allows the display of both live and stored data.
|
|
Hence, the objective MO2 is fulfilled.
|
|
|
|
The solution is able to send a notification when the room @co2 level reach the 1200 @ppm threshold.
|
|
However, the objective MO3 asks for a forecasting using a physical model.
|
|
Even if said physical model were studied during this project, as they are not included for the notification's sending, the objective
|
|
MO3 can not be considered as fulfilled.
|
|
|
|
Moreover, the forecasting used for the recommendation are based on the evolution of the @co2 level based on the amount of people in the room.
|
|
However, the outside temperature is not part of the analysis nor, incidentally, the airing recommendation.\
|
|
This absence is mainly explained by the team reduction.
|
|
The dropping member was initially tasked to implemented the notification system and the outside temperature retrieval.
|
|
While the MO3 was adjusted, an oversight from the team let this objective as unchanged.
|
|
As such, not enough effort were made available for the fulfilling of MO4.
|
|
|
|
Likewise, the forecasting could not be implemented thus forbidding the display of the data it would have provided.
|
|
As such, the optional objective OO1 could not be fulfilled.
|
|
|
|
Lastly, the time between the sending of two data point by a given node evolves based on the room recorded values.
|
|
Hence, the optional objective OO2 is fulfilled.
|
|
|
|
== Encountered difficulties
|
|
The project encountered some difficulties impacting the overall result.
|
|
While some are typically integration mishaps and is documented in @sec:impl, other are wider.
|
|
|
|
Namely, the team effort reduction following the dropout of a member impacted significantly the results.
|
|
As indicated in @sec:objectives, the client has accepted to adjust the project's objectives accordingly.
|
|
|
|
As warned by the client at the project's start, the actual room data arrived late in the project.
|
|
The solving of every observed issues lead to stable data available only two weeks prior to the project end.
|
|
This short time did not allow to give their full purpose to the data nor retrieve user feedback.
|
|
|
|
Lastly, the project still suffers from noticeable issues that are documented prior and in the GitHub issues.
|
|
|
|
== Future perspectives
|
|
While the project next iteration features shall be decided then, the current team recommend the following, without any wanted order:
|
|
- Deployment in every room at the Provence campus.
|
|
- Talk with the administration to send the notification on teams, since it is the standard communication medium.
|
|
- Build a dedicated board for the nodes, enabling the "low-power" that is sought for these elements.
|
|
- Implement a forecasting, including the outside temperature as intended at the very start of the project.
|
|
This forecasting shall include de room temperature as it also impact the people focus.
|
|
|
|
The github repository contains several opened issues tracking new features idea and observed issue to solve.
|
|
The next iteration's team is welcomed to consider these issues.
|
|
|
|
Furthermore, the following would improve the product usability, but requires dedicated development.
|
|
|
|
- Counting the amount of people in a room is paramount to have the physical model work properly.
|
|
Hence, having an automated way to count such shall be investigated.
|
|
A way to do so may be to have a camera and process the images.
|
|
This solution comes with high security around personal data protection.\
|
|
Another way that could be investigated is to estimate the amount of people from the @co2 level evolution.
|
|
A dedicated step in the deployment may enable efficient estimation.
|
|
- The retrieval on the window opening status could also be improved as the deployment remains very tinkered.
|
|
Dedicated casing could be designed and the room door may also be equipped.
|
|
Furthermore, a more general way could be investigated as the deployed one requires to have every window be equipped and by extension one node per window.
|
|
Moreover, a room may have hard-to-access window and thus could not be equipped easily.
|
|
- The @co2 level in a room with automated airing or special insulation may require modifications in the physical model.
|
|
These shall be investigated in a later iteration. |