Creating Speed Profile and daily profile table using HERE Traffic Pattern

[AbhishekPandey]

Hi,

I am using HERE Traffic Pattern and navstreet data to create a Historical traffic data and network for my analysis, but as mentioned in this guide: Historical traffic—ArcMap | Documentation (arcgis.com) under "How Network Analyst models historical traffic" my daily profile data is not in scale factor and hence I am getting a total green (freeflow) traffic map which is incorrect.

I have the Freeflow Speed (from Navstreet Feature class) and Travel Speed (HERE Traffic Pattern) for given links.

Now my question is How can I convert my daily profile to scale factors or do I need to create a new table? If so How can I do that?

I Have attached Screenshot of my Street Profile, Daily Profile and Spatial Join of navstreet-Street profile for better understanding

Any Help or guidance is appreciated

street profile (HERE Traffic)

Daily profile

StreetProfile_Navstreet Spatial Joined (as used in fgdb)

Thanks,

Abhishek

[AlanHatakeyama]

Hi Abhishek,

Yes, you do need to create a new table that has these scale factor values.  You will need to determine what is the maximum speed for each PATTERN_ID and then divide each speed in that PATTERN_ID row by that maximum.  Given that this table starts off as a .csv file, you can probably most easily do this in Microsoft Excel with function calculations within the cells, and then re-saving the result back to a .csv file before importing it into your geodatabase to include in the network dataset.

Note that each street references up to seven different PATTERN_ID values (one for each day of the week), so it may be possible that the maximum speed for each PATTERN_ID may not be the same.  You'll need to choose which daily maximum speed to use for your Freeflow Speed for that street -- maybe you take the most common maximum, or maybe you average the seven speeds, etc.  Usually when these maximum speeds are different, they only differ by ±1km/h, so the shift in travel time would be negligible -- especially since the original data seems to already be rounded to the nearest whole km/h anyway.

Please let me know if you have any further questions.  Thanks!

Alan

[AbhishekPandey · Accepted solution]

After weeks of searching, I finally settled with Pandas. In order to create the above-mentioned tables, I created inner joins, divided the values by actual speed to get speed-factor, and extracted new values in a different table with only unique values. Finally, I re-populated the existing table (pandas inner join) with the index of the newly created table.