estimate separate marginal util hour for bike

src/main/python/choicemodels/estimate_biogeme_plan_choice.py

@@ -33,6 +33,7 @@
     parser.add_argument("--est-price-perception-pt", help="Estimate price perception", action="store_true")
     parser.add_argument("--same-price-perception", help="Only estimate one fixed price perception factor", action="store_true")
     parser.add_argument("--price-perception", help="Given value for fixed price perception", type=float, default=1)
+    parser.add_argument("--effort", help="Additional time utility", nargs="+", action='append', default=[])
     parser.add_argument("--ascs", help="Predefined ASCs", nargs="+", action='append', default=[])
     parser.add_argument("--car-util", help="Fixed utility for car", type=float, default=None)
     parser.add_argument("--no-income", help="Don't consider the income", action="store_true")
@@ -59,6 +60,10 @@
     if fixed_ascs:
         print("Using fixed ascs", fixed_ascs)
 
+    effort = {x: float(y) for x, y in args.effort}
+    if effort:
+        print("Using time effort", effort)
+
     # Variables for constants
     ASC = {}
 
@@ -148,6 +153,10 @@
 
         u += v[f"plan_{i}_car_used"] * B_CAR
 
+        # Add the effort component (additional time disutility)
+        for mode, val in effort.items():
+            u -= v[f"plan_{i}_{mode}_hours"] * val
+
         if args.est_error_component:
             errs = 0
             for mode in ds.modes:
@@ -178,6 +187,8 @@
         modelName += "_util_money"
     if args.ascs:
         modelName += "_fixed_ascs"
+    if effort:
+        modelName += "_effort"
     if args.no_income:
         modelName += "_no_income"
     if args.est_price_perception_car:

src/main/python/choicemodels/estimate_biogeme_trip_choice.py

@@ -25,6 +25,7 @@
     parser.add_argument("--est-price-perception-pt", help="Estimate price perception", action="store_true")
     parser.add_argument("--est-ride-alpha", help="Estimate ride detour parameter", action="store_true")
     parser.add_argument("--est-bike-effort", help="Estimate parameter for bike effort", action="store_true")
+    parser.add_argument("--est-exp-dist", help="Modes for which to estimate distance elasticity", nargs="+", type=str, default=[])
     parser.add_argument("--same-price-perception", help="Only estimate one fixed price perception factor", action="store_true")
 
     parser.add_argument("--no-income", help="Don't consider the income", action="store_true")
@@ -39,6 +40,11 @@
     database = db.Database("data/choices", df)
     v = database.variables
 
+    mean_dist = df.groupby("choice").agg(dist=("beelineDist", "mean")) * 1000
+
+    print("Mean trip distance", (df.beelineDist * 1000).mean())
+    print("Mean trip distance by choice", mean_dist)
+
     ASC = {}
     for mode in ds.modes:
         # Base asc
@@ -66,7 +72,12 @@
     # THe detour factor for ride trip, influences the time costs, as well as distance cost
     BETA_RIDE_ALPHA = Beta('BETA_RIDE_ALPHA', 1, 0, 2, ESTIMATE if args.est_ride_alpha else FIXED)
 
-    BETA_BIKE_EFFORT = Beta('BETA_BIKE_UTIL_KM', 0, 0, 10, ESTIMATE if args.est_bike_effort else FIXED)
+    EXP_DIST = {}
+    for mode in args.est_exp_dist:
+        print(f"Estimating distance elasticity for {mode}")
+        EXP_DIST[mode] = (Beta(f'BETA_DIST_{mode}', 1, None, None, ESTIMATE), Beta(f'EXP_DIST_{mode}', 1, None, None, ESTIMATE))
+
+    BETA_BIKE_EFFORT = Beta('BETA_BIKE_UTIL_H', 0, 0, 10, ESTIMATE if args.est_bike_effort else FIXED)
 
     for i, mode in enumerate(ds.modes, 1):
         # Ride incurs double the cost as car, to account for the driver and passenger
@@ -82,6 +93,10 @@
         if mode == "bike":
             u -= v[f"{mode}_hours"] * BETA_BIKE_EFFORT
 
+        if mode in EXP_DIST:
+            beta, exp = EXP_DIST[mode]
+            u += beta * ((v[f"{mode}_km"] * 1000) / float(mean_dist.loc[i].dist)) ** exp
+
         U[i] = u
         AV[i] = v[f"{mode}_valid"]