diff --git a/Prosumer/scripts/time_series_processing.py b/Prosumer/scripts/time_series_processing.py
index 0b148087bd68e63fcbb770d2cdcb5b1f2c6b3c5b..32c63865a450b2cf5958cb5170a2f62d50d1a98d 100644
--- a/Prosumer/scripts/time_series_processing.py
+++ b/Prosumer/scripts/time_series_processing.py
@@ -135,19 +135,44 @@ def generate_profile(t_start, t_step, t_horizon, prediction='Perfect', **kwargs)
         elif profile == 'generate_therm_demand_tek':
             bld_typ = kwargs[profile][0]
             total_demand = kwargs[profile][1]
-            weather = kwargs[profile][2]
+            weather = kwargs[profile][2]['temperature'].values
             year = int(pd.Timestamp(t_start).year)
 
-            input_profiles['therm_demand'] = tek.gen_heat_profile(
-                building_typ=bld_typ, temperature_profile=weather, year=year,
-                yearly_demand=total_demand)
+            demands = tek.gen_heat_profile(building_typ=bld_typ,
+                                           temperature_profile=weather,
+                                           year=year,
+                                           yearly_demand=total_demand)
+
+            profile_date = pd.date_range(start=str(year)+"0101", periods=8760,
+                                         freq='H')
+            demands = pd.Series(demands, index=profile_date)
+
+            predictors['therm_demand'] = PredictionGenerator(demands,
+                                                             method=prediction)
+            input_profiles['therm_demand'] = predictors['therm_demand'].predict(
+                str(t_step) + 'H',
+                t_start,
+                t_horizon / t_step)
         elif profile == 'generate_water_demand_tek':
             bld_typ = kwargs[profile][0]
             total_demand = kwargs[profile][1]
             year = int(pd.Timestamp(t_start).year)
 
-            input_profiles['hot_water_demand'] = tek.gen_hot_water_profile(
+            # input_profiles['hot_water_demand'] = tek.gen_hot_water_profile(
+            #     building_typ=bld_typ, year=year, yearly_demand=total_demand)
+
+            demands = tek.gen_hot_water_profile(
                 building_typ=bld_typ, year=year, yearly_demand=total_demand)
+            profile_date = pd.date_range(start=str(year) + "0101", periods=8760,
+                                         freq='H')
+            demands = pd.Series(demands, index=profile_date)
+
+            predictors['hot_water_demand'] = PredictionGenerator(demands,
+                                                                 method=prediction)
+            input_profiles['hot_water_demand'] = predictors['hot_water_demand'].predict(
+                str(t_step) + 'H',
+                t_start,
+                t_horizon / t_step)
         else:
             # apply prediction methods upon other input profiles (air_temp, irradiance, etc.)
             predictors[profile] = PredictionGenerator(kwargs[profile], method=prediction)