Chris Mounts

Background

The traditional workflow for our scanners has always been:

1. Setup the scanner in position and adjust targets
2. Run a the scan
3. Select the targets
4. Acquire the targets

Repeat..

With the release of Leica Cyclone 7.2 came the ability to extract black and white targets from scan data automatically. This for the first time would allow us to skip steps 2 and 3 of our process and potentially see vast improvements to our data acquisition speed.
It would take several test and production trips for us to feel comfortable enough to trust in this new ability. Now almost a year into using auto-targeting I feel confident that the process has worked and successfully reduced our acquisition time driving down the overall hours to complete projects.

Disclaimer

The Challanges

Possibly the biggest drawback to this approach is missed targets. There is no avoiding the fact that on a project you will have missing targets. Sometimes it is the software that can’t seem to pull a target that is clearly visible, sometimes it’s a operator error of not making sure enough targets are visible in scan. Over the year we have seen improvements as our scan techs have adjusted to the new process however the bottom line is; there will be issues and they will have to be corrected in the office.

The Benefits

The process has delivered on our hopes of reducing field time. In fact were able to acquire data so quickly that we have a need to increase the number of targets we own to keep up with the movement of the scanner. So far the only trade off, for us, has been a slight increase in office time. Another benefit of the process is we have been able to greatly simplify the collection of data. By proxy this has increased our pool of scanner technicians.

The Results

For some time now I have worked by the principle of “You can not improve what you do not measure”. As part of this I have began measuring what I can and in this case have a sizable body of data covering hundreds of scans and many completed projects. From this data I have chosen two very similar projects that accurately represent the average numbers I have seen.

As shown, there was a dramatic drop in minutes per scan. These numbers take into account the entire work required for 20 positions. It includes things such as breaks and issues with the scanner into the average. As mentioned, this was very typical of the hundreds of scans I have data on. To see the full story you really need to look into things on a scan by scan basis.

Although each project varies in its specifics each follow this approximate pattern. That is to say that the auto-acquisition projects follow a relatively consistent path hovering around 9 minutes scan-to-scan with the occasional bump for a break or lunch. On the other hand field acquired projects almost always display an aggressive saw tooth pattern. They typically begin with several scans in the 20-25 minute range then start jumping around as targeting becomes more complex. Towards the end of the day targeting typically becomes easier and things begin to pick up again.

The data also reveals other interesting information, for example projects using field acquisition on average take more breaks and for longer times. It also shows the extreme impact that issues can have for example if a key target can not be acquired and a scan must be redone the effect can often push the scan-to-scan time to over an hour for field acquired while recovery for auto-acquired is only an additional 4-5 minutes.

The extended results of this workflow go far beyond scan-to-scan times as one might deduce. For projects requiring multi day travel the savings of expenses can be sizable.