VZD: Some records present incorrect speed_mgmt_points values

[atdservicebot]

Xavier noticed a few records not getting the correct values for “speed_mgmt_points”, this is a list of records with the reported problem:

14141018
14235350
14378407
14436288
14533327
14542851
14757828
14747769
15097928
14883007
14909386
15791258
15129084
15191354
15198132
15320483
15397272
15432109
15462005
15511421
15562524
15669405
15767058
16617630
16941430
16957071
17052450

Migrated from atd-vz-data #646

[patrickm02L]

@frankhereford figure out if there's still an existing problem. Make comments or recommend on closing.

[frankhereford]

@patrickm02L: I recommend that this issue be closed, as I find these crashes to have the correct speed management points calculated per crash.

As an aside, for planning purposes, I would like to note that there is some discussion among the speed management team about the scoring algorithm used, so it is not impossible that we'll be revisiting speed management points in the coming weeks or months, but that is not germane to this issue.

To this issue however, I have reviewed the bulk of these crashes, and the main theme is that they have a fatality, but it is marked as a non-ATD fatality. These are fatalities which do not meet one of these criteria:

• Death resulting from injuries sustained in crash within 30 days
• Crashes on Highways/freeways are included
• Crashes not occurring on private driveways/parking lots
• Crashes not involved in a police chase ("legal intervention")
• Crashes not involving a medical event that killed the driver before the collision
• Crashes not occurring from deliberate intent
  • Example: Suicide, homicide, injury, or damage purposefully inflicted
• Crashes that are not transportation road deaths
  • Example: Train colliding with another train at a road intersection

Upon consultation with @xavierapostol, he has confirmed that these non-ATD fatalities should not contribute to speed management points, which they currently do not. As things are working as expected, I believe that this issue was addressed at some point, but this issue slipped by without being closed. A mirror issue was closed in the VZ repo, as noted above.

[patrickm02L]

Thanks for the comprehensive overview @frankhereford. Thanks for flagging the speed management points scoring algorithm needs to be addressed in the future.