Merge pull request #127 from WISDEM/release-2.2.7

Release 2.2.7

CHANGELOG.md

@@ -53,10 +53,20 @@
 
 ## 2.2.5
 
-+ Fixed a solver problem that would hang during foundation calculations.
++ Fixed a solver problem that would hang during foundation calculations. This was a bug encountered in the upgrade to SymPy 1.5.1
 
 + Added post processing scripts to load data into PostgreSQL
 
 ## 2.2.6
 
 + Fix setup.py to automatically find the `landbosse` package.
+
+## 2.2.7
+
++ Chnaged construction duration functionality.
+
++ Weather window can now be extended to an arbitrarily long length by duplicating the underlying wind data.
+
++ All modules now report the actual construction time they require to perform the scope of work they model.
+
++ ManagementCost now keeps the management crew onsite for only the time necessary to complete all scope of work.

landbosse/excelio/WeatherWindowCSVReader.py

@@ -1,10 +1,14 @@
+from math import ceil
+
 import pandas as pd
 
+
 SEASON_WINTER = 'winter'
 SEASON_SPRING = 'spring'
 SEASON_SUMMER = 'summer'
 SEASON_FALL = 'fall'
 
+
 month_numbers_to_seasons = {
     1: SEASON_WINTER,
     2: SEASON_WINTER,
@@ -101,12 +105,14 @@ def read_weather_window(weather_data, local_timezone='America/Denver'):
     # Convert UTC to local time
     weather_data['Date'] = weather_data['Date UTC'].dt.tz_convert(local_timezone)
 
-    # Extract year, month, day, hour from the local date
-    weather_data['Year'] = weather_data['Date'].dt.year
+    # Extract month, day, hour from the local date
     weather_data['Month'] = weather_data['Date'].dt.month
     weather_data['Day'] = weather_data['Date'].dt.day
     weather_data['Hour'] = weather_data['Date'].dt.hour
 
+    # The original date columns are now redundant. Drop them.
+    weather_data.drop(columns=['Date', 'Date UTC'])
+
     # create time window for normal (8am to 6pm) versus long (24 hour) time window for operation
     weather_data['Time window'] = weather_data['Hour'].between(8, 18, inclusive=True)
     boolean_dictionary = {True: 'normal', False: 'long'}
@@ -122,3 +128,51 @@ def read_weather_window(weather_data, local_timezone='America/Denver'):
 
     # return the result
     return weather_data
+
+def extend_weather_window(weather_window_df, months_of_weather_data_needed):
+    """
+    This function extends a weather window by duplicating the rows to create
+    a weather window that spans the needed number of months.
+
+    Suppose that a weather window is 8760 hours long, but that 72 months
+    (52560 hours) are needed. This method will create a new dataframe
+    that has the original 8760 hours duplicated 6 times.
+
+    However, if the number of needed months is less than or equal to the
+    duration of the weather window, a reference to the original weather
+    window is returned. Also, the number of rows returned will always be
+    a multiple of the number of rows in the original data frame.
+
+    If rows are added to the weather window, they are added to a new dataframe.
+    The weather window is not modified in place.
+
+    Parameters
+    ----------
+    weather_window_df : pd.DataFrame
+        The original weather window.
+
+    months_of_weather_data_needed : int
+        The number of months of weather data needed. Each month is approximated
+        to have 730 hours.
+
+    Returns
+    -------
+    pd.DataFrame
+        If the weather window accommodates the necessary number of months, then
+        a reference to the original dataframe is returned. Otherwise, a reference
+        to a new dataframe that has a row count that is a multiple of the number
+        of rows in the original weather window.
+    """
+    hours_per_month = 730
+    hours_of_weather_data_needed = hours_per_month * months_of_weather_data_needed
+    hours_of_weather_data_available = len(weather_window_df)
+
+    if hours_of_weather_data_needed <= hours_of_weather_data_available:
+        return weather_window_df
+
+    number_of_windows_needed = int(ceil(hours_of_weather_data_needed / hours_of_weather_data_available))
+    weather_window_as_list = weather_window_df.to_dict(orient='records')
+    weather_window_repeated_as_list = weather_window_as_list * number_of_windows_needed
+    result = pd.DataFrame(weather_window_repeated_as_list)
+
+    return result

landbosse/excelio/XlsxFileOperations.py

@@ -48,7 +48,7 @@ def get_input_output_paths_from_argv_or_env(self):
 
         There is a fourth option this method looks for:
 
-        --scaling-study or -s
+        --scaling or -s
 
         This enables auto scaling of input parameters as found in XlsxReader's
         apply_cost_and_scaling_modifications_to_project_parameters() method.
@@ -81,7 +81,7 @@ def get_input_output_paths_from_argv_or_env(self):
         validation_enabled =  '--validate' in sys.argv or '-v' in sys.argv
 
         # This is for scaling study operation
-        enable_scaling_study = '--scaling-study' in sys.argv or '-s' in sys.argv
+        enable_scaling_study = '--scaling' in sys.argv or '-s' in sys.argv
 
         # If validation and scaling study are simultaneously enabled, raise
         # an error

landbosse/excelio/XlsxReader.py

@@ -5,7 +5,7 @@
 from math import ceil
 
 from .XlsxOperationException import XlsxOperationException
-from .WeatherWindowCSVReader import read_weather_window
+from .WeatherWindowCSVReader import read_weather_window, extend_weather_window
 from ..model import DefaultMasterInputDict
 from .GridSearchTree import GridSearchTree
 
@@ -392,12 +392,6 @@ def create_master_input_dictionary(self, project_data_dataframes, project_parame
         incomplete_input_dict['site_facility_building_area_df'] = project_data_dataframes['site_facility_building_area']
         incomplete_input_dict['material_price'] = project_data_dataframes['material_price']
 
-        # The weather window is stored on a sheet of the project_data, but
-        # needs preprocessing after it is read. The preprocessing changes it
-        # from wind toolkit format to a dataframe.
-        weather_window_input_df = project_data_dataframes['weather_window']
-        incomplete_input_dict['weather_window'] = read_weather_window(weather_window_input_df)
-
         # Read development tab, if it exists (it is optional since development costs can
         # be placed in the project list):
         if 'development' in project_data_dataframes:
@@ -500,6 +494,15 @@ def create_master_input_dictionary(self, project_data_dataframes, project_parame
             incomplete_input_dict['markup_overhead'] = project_parameters['Markup overhead']
             incomplete_input_dict['markup_profit_margin'] = project_parameters['Markup profit margin']
 
+        # The weather window is stored on a sheet of the project_data, but
+        # needs preprocessing after it is read. The preprocessing changes it
+        # from wind toolkit format to a dataframe.
+        number_of_months_for_construction = int(project_parameters['Total project construction time (months)'])
+        weather_window_input = project_data_dataframes['weather_window']
+        weather_window_intermediate = read_weather_window(weather_window_input)
+        extended_weather_window = extend_weather_window(weather_window_intermediate, number_of_months_for_construction)
+        incomplete_input_dict['weather_window'] = extended_weather_window
+
         # Now fill any missing values with sensible defaults.
         defaults = DefaultMasterInputDict()
         master_input_dict = defaults.populate_input_dict(incomplete_input_dict=incomplete_input_dict)
@@ -633,7 +636,9 @@ def create_serial_number(self, project_id, index, max_index):
         total_digit_count = 1
         index_digit_count = len(str(index))
 
-        if 0 < max_index < 1e1 - 1:
+        if max_index < 10:
+            total_digit_count = 1
+        elif 0 < max_index < 1e1 - 1:
             total_digit_count = 1
         elif 1e1 <= max_index < 1e2 - 1:
             total_digit_count = 2

landbosse/model/CollectionCost.py

@@ -653,14 +653,14 @@ def estimate_construction_time(self, construction_time_input_data, construction_
         boolean_dictionary = {True: collection_construction_time * 30, False: np.NAN}
         operation_data['time_construct_bool'] = operation_data['time_construct_bool'].map(boolean_dictionary)
         operation_data['Time construct days'] = operation_data[['time_construct_bool', 'Number of days taken by single crew']].min(axis=1)
-        num_days = operation_data['Time construct days']
+        num_days = operation_data['Time construct days'].max()
 
         # pull out management data
         crew_cost = self.input_dict['crew_cost']
         crew = self.input_dict['crew'][self.input_dict['crew']['Crew type ID'].str.contains('M0')]
         management_crew = pd.merge(crew_cost, crew, on=['Labor type ID'])
-        management_crew = management_crew.assign(per_diem_total=management_crew['Per diem USD per day'] * management_crew['Number of workers'] * num_days.iloc[0])
-        management_crew = management_crew.assign(hourly_costs_total=management_crew['Hourly rate USD per hour'] * self.input_dict['hour_day'][self.input_dict['time_construct']]  * num_days.iloc[0])
+        management_crew = management_crew.assign(per_diem_total=management_crew['Per diem USD per day'] * management_crew['Number of workers'] * num_days)
+        management_crew = management_crew.assign(hourly_costs_total=management_crew['Hourly rate USD per hour'] * self.input_dict['hour_day'][self.input_dict['time_construct']] * num_days)
         management_crew = management_crew.assign(total_crew_cost_before_wind_delay=management_crew['per_diem_total'] + management_crew['hourly_costs_total'])
         self.output_dict['management_crew'] = management_crew
         self.output_dict['managament_crew_cost_before_wind_delay']= management_crew['total_crew_cost_before_wind_delay'].sum()
@@ -915,8 +915,13 @@ def run_module(self):
             self.weather_input_dict[
                 'wind_height_of_interest_m'] = self.input_dict['critical_height_non_erection_wind_delays_m']
 
-            # compute and specify weather delay mission time for roads
+            # Compute the duration of the construction for electrical collection
             duration_construction = operation_data['Time construct days'].max(skipna=True)
+            days_per_month = 30
+            duration_construction_months = duration_construction / days_per_month
+            self.output_dict['collection_construction_months'] = duration_construction_months
+
+            # compute and specify weather delay mission time for roads
             operational_hrs_per_day = self.input_dict['hour_day'][self.input_dict['time_construct']]
             mission_time_hrs = duration_construction * operational_hrs_per_day
             self.weather_input_dict['mission_time_hours'] = int(mission_time_hrs)