Reconstruct data

*This video does not contain an audio track 

 

Content: short-term data anomalies, advanced data management, automatic data reconstruction tool

Prerequisites: data anomaly/gap under 7 days, 4 weeks of good data preceding the first day of the anomaly, good data after the anomaly

Context: The continuity of data collection can sometimes be interrupted: cycle track closed for works, mobile counter moved, batteries discharged, etc. Our automatic reconstruction tool allows you to reconstruct single days, up to 7 consecutive days.

To reconstruct a day of the week, our tool will calculate hourly averages of the data collected on that day during the previous 4 weeks.

Example: To reconstruct a Sunday, the new data will be based on the averages of 4 Sundays before that.

You can only use this feature to correct up to 7 consecutive days of anomalous data!

 

Step-by-step instructions:

We recommend to always back up your data before performing any data manipulations in Eco-Visio. Here is our guide on how to export data from a counting site!

 

Step 1: Go to the Analysis page and visualize the anomalous time period. You can use the zoom in feature by clicking the left mouse button, holding, and moving it to highlight the period in question

Step 2: Under the three dots Action menu you will find ‘Edit Data‘ feature. Click to launch the Data Editing window

Step 3: Select the first anomalous day by leveraging arrow buttons located in the top middle of the data graph -> Click on ‘Auto Reconstruction’. Note that you can also reconstruct separate flows by selecting the anomalous one in the upper left corner of the data editing window

Step 4: Validate the reconstructed data and repeat for all other affected dates

 

P.S. If your data gap is longer than 7 consecutive days, please contact our team of Client Consultants (sales@eco-counter.com) so they could engage Eco-Counter data reconstruction experts, who use an advanced regression algorithm to estimate the missing counts.

 

Happy reconstructing!

Related articles