Use new NTD monthly data sources for Staging and

Mart tables - Also Format SQL files and reorder columns - Remove irrelevant year column from dim model - Filter out null NTD_ids - Edit description on documentation [#3519]
cal-itp · Nov 13, 2024 · f77bf8e · f77bf8e
1 parent 142231c
commit f77bf8e
Show file tree

Hide file tree

Showing 16 changed files with 497 additions and 446 deletions.
diff --git a/script/scrape_ntd.py b/script/scrape_ntd.py
@@ -24,7 +24,9 @@
 poetry install
 then
 poetry run python scrape_ntd.py annual-database-agency-information 2021 https://www.transit.dot.gov/sites/fta.dot.gov/files/2022-10/2021%20Agency%20Information.xlsx
-poetry run python scrape_ntd.py monthly-ridership-with-adjustments 2024 https://www.transit.dot.gov/sites/fta.dot.gov/files/2024-04/February%202024%20Complete%20Monthly%20Ridership%20%28with%20adjustments%20and%20estimates%29_240402_0.xlsx
+
+-- poetry run python scrape_ntd.py monthly-ridership-with-adjustments 2024 https://www.transit.dot.gov/sites/fta.dot.gov/files/2024-04/February%202024%20Complete%20Monthly%20Ridership%20%28with%20adjustments%20and%20estimates%29_240402_0.xlsx
+-- REPLACED by "/airflow/plugins/operators/scrape_ntd_xlsx.py"
 """
 
 import gzip

diff --git a/warehouse/models/docs/_docs_ntd.md b/warehouse/models/docs/_docs_ntd.md
@@ -81,6 +81,10 @@ A system for carrying transit passengers described by specific right-of-way (ROW
     - Jitney (JT)    
 {% enddocs %}
 
+{% docs ntd_3_mode %}
+A grouping of modes based upon whether the mode operates on rail, is a bus mode, is ferry boat service or other.
+{% enddocs %}
+
 {% docs ntd_time_period %}
 The time period for which data was collected.
 Valid values are: 
@@ -124,3 +128,52 @@ Actual vehicle hours exclude:
     - Operator training,
     - Vehicle maintenance testing.
 {% enddocs %}
+
+{% docs ntd_monthly_upt %}
+Unlinked Passenger Trips (UPT) -
+The number of passengers who board public transportation vehicles.
+Passengers are counted each time they board vehicles no matter how
+many vehicles they use to travel from their origin to their destination.
+{% enddocs %}
+
+{% docs ntd_monthly_voms %}
+Vehicles Operated in Annual Maximum Service (VOMS) -
+The number of revenue vehicles operated to meet the annual maximum
+service requirement. This is the revenue vehicle count during the peak
+season of the year; on the week and day that maximum service is
+provided. Vehicles operated in maximum service (VOMS) exclude:
+    - Atypical days; or
+    - One-time special events.
+{% enddocs %}
+
+{% docs ntd_monthly_vrh %}
+Vehicle Revenue Hours (VRH) -
+The hours that vehicles are scheduled to or actually travel while in
+revenue service. Vehicle revenue hours include:
+    - Layover / recovery time.
+Vehicle revenue hours exclude:
+    - Deadhead;
+    - Operator training;
+    - Vehicle maintenance testing; and
+    - Other non-revenue uses of vehicles.
+{% enddocs %}
+
+{% docs ntd_monthly_vrm %}
+Vehicle Revenue Miles (VRM) -
+The miles that vehicles are scheduled to or actually travel while in
+revenue service. Vehicle revenue miles include:
+    - Layover / recovery time.
+Vehicle revenue miles exclude:
+    - Deadhead;
+    - Operator training;
+    - Vehicle maintenance testing; and
+    - Other non-revenue uses of vehicles.
+{% enddocs %}
+
+{% docs ntd_monthly_mode_type_of_service_status %}
+Indicates whether a property reports (active) or not (inactive) during the most recent Annual report year.
+{% enddocs %}
+
+{% docs ntd_monthly_service_type %}
+A summarization of modes into `Fixed Route`, `Demand Response`, or `Unknown`.
+{% enddocs %}
diff --git a/warehouse/models/intermediate/ntd/int_ntd.yml b/warehouse/models/intermediate/ntd/int_ntd.yml
@@ -1,27 +1,27 @@
 version: 2
+
 models:
     - name: int_ntd__monthly_ridership_with_adjustments_upt
-      description: |
-        Ridership - upt
+      description: '{{ doc("ntd_monthly_upt") }}'
       config:
         materialized: table
+
     - name: int_ntd__monthly_ridership_with_adjustments_vrm
-      description: |
-        Ridership - vrm
+      description: '{{ doc("ntd_monthly_vrm") }}'
       config:
         materialized: table
+
     - name: int_ntd__monthly_ridership_with_adjustments_vrh
-      description: |
-        Ridership - vrh
+      description: '{{ doc("ntd_monthly_vrh") }}'
       config:
         materialized: table
+
     - name: int_ntd__monthly_ridership_with_adjustments_voms
-      description: |
-        Ridership - voms
+      description: '{{ doc("ntd_monthly_voms") }}'
       config:
         materialized: table
+
     - name: int_ntd__monthly_ridership_with_adjustments_joined
-      description: |
-        Ridership - joined
+      description: Ridership - joined VRM, VOMS, VRH, and UPT
       config:
         materialized: table
diff --git a/warehouse/models/intermediate/ntd/int_ntd__monthly_ridership_with_adjustments_joined.sql b/warehouse/models/intermediate/ntd/int_ntd__monthly_ridership_with_adjustments_joined.sql
@@ -1,60 +1,57 @@
 {{ config(materialized="table") }}
 
-with
-    voms as (
-        select * from {{ ref("int_ntd__monthly_ridership_with_adjustments_voms") }}
+WITH
+    voms AS (
+        SELECT * FROM {{ ref("int_ntd__monthly_ridership_with_adjustments_voms") }}
     ),
 
-    vrh as (select * from {{ ref("int_ntd__monthly_ridership_with_adjustments_vrh") }}),
-    vrm as (select * from {{ ref("int_ntd__monthly_ridership_with_adjustments_vrm") }}),
-    upt as (select * from {{ ref("int_ntd__monthly_ridership_with_adjustments_upt") }}),
+    vrh AS (SELECT * FROM {{ ref("int_ntd__monthly_ridership_with_adjustments_vrh") }}),
+    vrm AS (SELECT * FROM {{ ref("int_ntd__monthly_ridership_with_adjustments_vrm") }}),
+    upt AS (SELECT * FROM {{ ref("int_ntd__monthly_ridership_with_adjustments_upt") }}),
 
-    int_ntd__monthly_ridership_with_adjustments_joined as (
-        select voms.*, upt.upt, vrm.vrm, vrh.vrh
-        from voms
+    int_ntd__monthly_ridership_with_adjustments_joined AS (
+        SELECT voms.*,
+               upt.upt,
+               vrm.vrm,
+               vrh.vrh
+          FROM voms
 
-        full outer join
-            upt
-            on voms.ntd_id = upt.ntd_id
-            and voms.year = upt.year
-            and voms.mode = upt.mode
-            and voms.reporter_type = upt.reporter_type
-            and voms.agency = upt.agency
-            and voms._3_mode = upt._3_mode
-            and voms.period_month = upt.period_month
-            and voms.period_year = upt.period_year
-            and voms.tos = upt.tos
-            and voms.mode_type_of_service_status = upt.mode_type_of_service_status
+          FULL OUTER JOIN upt
+            ON voms.ntd_id = upt.ntd_id
+           AND voms.mode = upt.mode
+           AND voms.reporter_type = upt.reporter_type
+           AND voms.agency = upt.agency
+           AND voms._3_mode = upt._3_mode
+           AND voms.period_month = upt.period_month
+           AND voms.period_year = upt.period_year
+           AND voms.tos = upt.tos
+           AND voms.mode_type_of_service_status = upt.mode_type_of_service_status
 
-        full outer join
-            vrm
-            on voms.ntd_id = vrm.ntd_id
-            and voms.year = vrm.year
-            and voms.mode = vrm.mode
-            and voms.reporter_type = vrm.reporter_type
-            and voms.agency = vrm.agency
-            and voms._3_mode = vrm._3_mode
-            and voms.period_month = vrm.period_month
-            and voms.period_year = vrm.period_year
-            and voms.tos = vrm.tos
-            and voms.mode_type_of_service_status = vrm.mode_type_of_service_status
+          FULL OUTER JOIN vrm
+            ON voms.ntd_id = vrm.ntd_id
+           AND voms.mode = vrm.mode
+           AND voms.reporter_type = vrm.reporter_type
+           AND voms.agency = vrm.agency
+           AND voms._3_mode = vrm._3_mode
+           AND voms.period_month = vrm.period_month
+           AND voms.period_year = vrm.period_year
+           AND voms.tos = vrm.tos
+           AND voms.mode_type_of_service_status = vrm.mode_type_of_service_status
 
-        full outer join
-            vrh
-            on voms.ntd_id = vrh.ntd_id
-            and voms.year = vrh.year
-            and voms.mode = vrh.mode
-            and voms.reporter_type = vrh.reporter_type
-            and voms.agency = vrh.agency
-            and voms._3_mode = vrh._3_mode
-            and voms.period_month = vrh.period_month
-            and voms.period_year = vrh.period_year
-            and voms.tos = vrh.tos
-            and voms.mode_type_of_service_status = vrh.mode_type_of_service_status
+          FULL OUTER JOIN vrh
+            ON voms.ntd_id = vrh.ntd_id
+           AND voms.mode = vrh.mode
+           AND voms.reporter_type = vrh.reporter_type
+           AND voms.agency = vrh.agency
+           AND voms._3_mode = vrh._3_mode
+           AND voms.period_month = vrh.period_month
+           AND voms.period_year = vrh.period_year
+           AND voms.tos = vrh.tos
+           AND voms.mode_type_of_service_status = vrh.mode_type_of_service_status
     -- where voms.ntd_id not in ("10089", "20170", "30069", "90178", "90179")
     -- These agencies have null for uace_cd and uza_name and perhaps are not good to
     -- have in the dataset.
     -- If you don't want them then add that where clause back in.
     )
-select *
-from int_ntd__monthly_ridership_with_adjustments_joined
+
+SELECT * FROM int_ntd__monthly_ridership_with_adjustments_joined
diff --git a/warehouse/models/intermediate/ntd/int_ntd__monthly_ridership_with_adjustments_upt.sql b/warehouse/models/intermediate/ntd/int_ntd__monthly_ridership_with_adjustments_upt.sql
@@ -1,72 +1,67 @@
-with
-    source_pivoted as (
+WITH
+    source_pivoted AS (
         {{
             dbt_utils.unpivot(
                 cast_to="int",
-                relation=ref("stg_ntd__monthly_ridership_with_adjustments_upt"),
+                relation=ref("stg_ntd_ridership_historical__complete_monthly_ridership_with_adjustments_and_estimates__upt"),
                 exclude=[
-                    "uza_name",
-                    "uace_cd",
-                    "dt",
-                    "ts",
-                    "year",
-                    "ntd_id",
-                    "reporter_type",
+                    "_3_mode",
                     "agency",
-                    "mode_type_of_service_status",
+                    "legacy_ntd_id",
                     "mode",
-                    "_3_mode",
+                    "mode_type_of_service_status",
+                    "ntd_id",
+                    "reporter_type",
                     "tos",
-                    "legacy_ntd_id",
+                    "uace_cd",
+                    "uza_name",
+                    "dt",
+                    "execution_ts"
                 ],
                 field_name="period",
                 value_name="upt",
             )
         }}
     ),
-    int_ntd__monthly_ridership_with_adjustments_upt as (
-        select
-            uza_name,
-            format("%05d", cast(uace_cd as int64)) as uace_cd,
-            dt as _dt,
-            ts,
-            year,
-            format("%05d", cast(ntd_id as int64)) as ntd_id,
-            legacy_ntd_id,
-            reporter_type,
-            agency,
-            mode_type_of_service_status,
-            mode,
-            _3_mode,
-            tos,
-            split(period, '_')[offset(2)] as period_year,
-            split(period, '_')[offset(1)] as period_month,
-            upt
-        from source_pivoted
-        where
-            mode in (
-                "DR",
-                "FB",
-                "LR",
-                "MB",
-                "SR",
-                "TB",
-                "VP",
-                "CB",
-                "RB",
-                "CR",
-                "YR",
-                "MG",
-                "MO",
-                "AR",
-                "TR",
-                "HR",
-                "OR",
-                "IP",
-                "AG",
-                "PB",
-                "CC"
-            )
+
+    int_ntd__monthly_ridership_with_adjustments_upt AS (
+        SELECT format("%05d", cast(cast(ntd_id AS NUMERIC) AS INT64)) AS ntd_id,
+               legacy_ntd_id,
+               agency,
+               reporter_type,
+               split(period, '_')[offset(2)] AS period_year,
+               split(period, '_')[offset(1)] AS period_month,
+               uza_name,
+               format("%05d", cast(uace_cd AS INT64)) AS uace_cd,
+               mode,
+               mode_type_of_service_status,
+               _3_mode,
+               tos,
+               upt,
+               dt AS _dt,
+               execution_ts
+          FROM source_pivoted
+         WHERE mode IN ("AG",
+                        "AR",
+                        "CB",
+                        "CC",
+                        "CR",
+                        "DR",
+                        "FB",
+                        "HR",
+                        "IP",
+                        "LR",
+                        "MB",
+                        "MG",
+                        "MO",
+                        "OR",
+                        "PB",
+                        "RB",
+                        "SR",
+                        "TB",
+                        "TR",
+                        "VP",
+                        "YR")
     )
-select *
-from int_ntd__monthly_ridership_with_adjustments_upt
+
+SELECT * FROM int_ntd__monthly_ridership_with_adjustments_upt