Merge pull request #198 from Caliper-Corporation/dev

Bring final model changes into main

docs/R/ieei.R

@@ -4,7 +4,7 @@ data_dir <- "data/input/ieei/"
 output_dir <- "data/output/ieei/"
 master_dir <- "../master/"
 
-taz_shape_filename <- paste0(master_dir, "tazs/master_tazs.shp")
+taz_shape_filename <- "data/input/tazs/master_tazs.shp"
 socec_filename <- paste0(data_dir, "se_2016.csv")
 consolidated_streetlight_filename <- paste0(private_dir, "streetlight-ieei-flows.RDS")
 ncstm_filename <- paste0(data_dir, "ncstm-demand.rds")

docs/_site.yml

@@ -31,6 +31,8 @@ navbar:
           href: parking.html
         - text: "Destination Choice"
           href: resident_dest_choice.html
+        - text: "Mode Choice"
+          href: resident_mode_choice.html
         - text: "Non-homebased"
           href: nhb.html
     - text: Other Models

docs/assignment_roadway.Rmd

@@ -71,54 +71,32 @@ added to HOV3+.
 
 ## Value of time
 
-Once trips are collapsed into the primary vehicle classes, they are split into
-different value of time categories. The procedure for doing this is borrowed
-from the NC Statewide Model
-([documentation](other_docs/NCSTM_TollDocumentation_Section11-2.docx)). For
-residents, this stratification is informed by household income, time of day of
-the trip, and by the length of the trip. 
+Once trips are collapsed into the primary vehicle classes, they are assigned
+values of time. The base values were borrowed from the NC Statewide Model, but
+were then calibrated to match NC 540 traffic counts.
 
-Qualitatively, value of time goes up:
-
-- As income increases
-- As trip length increases
-- In the peak periods (where work trips are most common)
-
-The NCSTM/TRMG2 procedure divides each class into five VOT categories, except
-SUT which is divided into three. These are shown in the table below.
 ```{r, include=FALSE}
 tbl <- read_csv("../master/assignment/vot_params.csv")
 ```
 
 ```{r}
-tbl <- tbl[1:18, c("Name", "Value")]
 tbl <- tbl %>%
-  separate(Name, into = c("a", "b", "c"), fill = "right") %>%
-  mutate(
-    Name = ifelse(!is.na(c), paste(a, b), a),
-    VOT = ifelse(is.na(c), b, c)
-  ) %>%
-  select(Name, VOT, Value) %>%
-  pivot_wider(names_from = Name, values_from = Value) %>%
-  mutate(VOT = gsub("vot", "", VOT)) %>%
-  rename(
-    `VOT Group` = VOT,
-    `PK Auto` = `pk auto`,
-    `OP Auto` = `op auto`,
-    SUT = sut,
-    MUT = mut
-  ) 
+  select(Name, Value) %>%
+  mutate(Name = case_when(
+    Name == "pk_auto" ~ "Autos/CV (PK)",
+    Name == "op_auto" ~ "Autos/CV (OP)",
+    Name == "sut" ~ "SUT",
+    Name == "mut" ~ "MUT",
+    TRUE ~ "Calibration Factor"
+  )) %>%
+  rename(`VOT ($/hr)` = Value)
+  
 
 tbl %>%
   kable() %>%
   kable_styling(full_width = FALSE)
 ```
 
-Such a high number of classes significantly increases assignment run time. To
-mitigate this, categories 1-3 for the auto classes are collapsed and use the
-values from VOT group 2. This trade off is reasonable given that it is the
-higher VOT categories that are most likely to consider toll facilities.
-
 ## Assignment parameters
 
 Relevant assignment details are listed below: