Uncategorized

Automation of Time and Motion Studies – Beta release

When Watch&Time App started, the focus was to bring a century old technique into the 21st century.   Taking into account the advances in operating systems, computing power and memory in mobile devices in the past few years, there is an opportunity to bring automation into play.  In effect, automating and streamlining an industrial engineering technique.

If you want to see what is possible, you can now download the beta:    https://testflight.apple.com/join/cuqu2YCN. Available for iPhone, iPad and Mac with M1 chips.

Some observations regarding on device video activity analysis.

Firstly, it is really CPU and memory intensive.  And therefore, not recommended for older devices.  There is a quite.a bit of optimisation done to be able to run it on an iPhone and iPad.  And this impact accuracy of the model.   There is still development being done to leverage the newest Apple devices with 8 or 16GB memory which will lead to more accuracy. Speaking about accuracy.   The machine learning model is still actively being developed.  Therefore, the automation is more of a first run pass on a video as the best accuracy at present with video models are only around 70-80%.   This will become better as better architectures are developed.

Secondly, the Apple microchips are really fast with machine learning predictions.  A 2019 iPad Pro outperforms the MacBook Pro Intel by factor of 2-3x.   Which is insane.  As we haven’t tested it on a M1 Ultra, there is an expected higher performance.  If anyone has a new Mac Studio, please download the beta and test.

Thirdly.  There has been a few requests for an Android version of Watch&Time.   At present, there is active development, however, with the advent of Windows 11 that will run Android, this is definitely being investigated as it will open the app to more devices and users.

So, what is next?   Logically, we need to optimise the model and the code.   It is now to bring it to a level that it is accurate enough for daily use.   We are also looking at bringing multiple options for analysing the video, meaning on device or via the Cloud.  Lastly, we are also looking at the ability to have your data and results shared across devices.   If you do the analysis on your new Mac Studio, it would be great to see the results from your phone after it is done.

Lastly.   We do need help on making the ML model more accurate.  So if you want the app for free (not the beta) and willing to share tagged videos of time and motion studies, please contact chris@watchtimeapp.com

Chris

Posted by boschris007, 0 comments

Next Update Incoming

Watch&Time will receive an update this week.  The normal bug fixes and optimisations are per usual.  The biggest change is the update is to ensure that it is compatible with iOS15.  During the beta of iOS15, there were some UI issues that had to be resolved.  So it is highly recommended to update accordingly.  The Mac app version of Watch&Time is currently being tested for the latest MacOS and will be updated just before the release.

Feedback is always welcome via Twitter, LinkedIn or sending an email to social@watchtimeapp.com.

C

Posted by boschris007, 0 comments

A brief history of Watch&Time

How it started and where it is going

Watch&Time app started in 2017 after one of my junior industrial engineers had to complete a time and motion study in a service company.  The engineer took a clipboard and paper and went off on a two week analysis. My team at that time had a coffee chat on why haven’t industrial engineers optimised one of the key techniques used for decades.  From Gilbreth to Taylor , not much has changed to the basic technique, even though it is still the best method to get an understanding standard times of activities.  After the discussion, I decided to build a basic time study app for the team.  It was never used, but a large consulting company needed exactly the same tool, so we tested the viability of the app on the project.

So, I set off the build an app with a set of comprehensive features to optimise time and motion studies.  It started off as an iPhone app with the classic time study, activity sampling and also a quick count feature to determine throughput.   But there was still something missing.   Why waste an engineer’s time standing on a workfloor, call centre etc. Why not optimise the way we do time and motion studies?  Watch&Time (and many other similar apps) only removed the need for the clipboard and pen, therefore only making a physical process, digital.  And the biggest benefit is no need to recapture the data.   So, what came next?

Video analysis.  What if you can do time and motion studies on videos. Watch&Time was released with the feature to complete time and motion studies on videos.  Either record the video with your iPhone or iPad, or import a video.   The biggest gains are two fold:  1) the engineer does not have to be present on the floor, therefore allowing them to work on other tasks and 2) accelerated analysis.   The latter makes it possible to complete an activity sampling (Random or interval) of an 8 (eight) hour video in 5 minutes, taking 120 samples.  Timestudy can be done at 2x, 4x or even 8x, bringing down the analysis time for an 8 (eight) hour video to just over an hour.  Quality of the data is also improved as the analyst can go back and not miss a task / activity.  And the best part, you can do it on your iPhone and iPad…..and Mac.

Even though the original focus of Watch&Time was to bring time and motion studies to mobile devices, video analysis created a new requirement for users to be able to complete the video from their laptops.   So, from ver 1.4, Watch&Time became available on the Mac via die Mac App Store, with all the same features available on the mobile devices.

Watch&Time is currently on ver 1.5 and the team is busy working hard on developing new features to even further optimise the way we do time and motion studies.  Do you want to join the Beta Program?  Email social@watchtimeapp.com to join.

Feedback and input is always welcome.

C

 

 

 

Posted by boschris007, 2 comments