Model Results. To explore the public’s AV adoption and WTP behavior, an SEM approach was employed. The measurement portion of the model is presented in Table 133. It revealed four major latent factors. Accordingly, Factor 1 is positively associated with the joy of driving and individuals’ unwillingness to multitask. Factor 2 indicates how individuals count on the utility they obtain from their mobility choices, ranging from monetary and time costs to functionality and convenience. Hence, we label this factor as “mode choice reasoning”. Factor 3 expresses individuals’ trust issues when using shared mobility. Finally, Factor 4 represents a combination of multitasking and technology savviness. Interestingly, attitudes related to benefits/concerns of private ownership did not show any significant contribution to the model. Table 13 Result of Measurement Equations for AV Adoption and WTP Attitudinal Questions Factor 1 Factor 2 Factor 3 Factor 4 Joy of driving Choice reasoning Trust issue Technology savviness Attitudes: Overall Perception of Shared Mobility Attitudes: Trust I believe that shared transportation services can help me save on my expenses On-demand services increase the quality of life I hardly trust to travel with strangers Traveling by myself (or with people I know) is much more convenient than with strangers I prefer doing one thing at a Coeff. Z Coeff. Z Coeff Z Coeff. Z 1.44 11.13 1.43 11.4
Model Results. An SEM approach was employed to estimate the expected time and cost saving to switch to ridesourcing. This section discusses the results of the measurement portion of the model, as presented in 15. It reveals seven major latent factors. Accordingly, Factor 1 is positively associated with the joy of driving and individuals’ unwillingness to multitask. Factor 2 indicates how individuals count on the utility they obtain from their mobility choices, ranging from monetary and time costs to functionality and convenience. Hence, we label this factor as “mode choice reasoning”. Factor 3 expressed individuals’ trust issues when using shared mobility. Finally, Factor 4 represents a combination of multitasking and technology savviness. Factors 5 through 7 represent the perception of private ownership advantages and disadvantages. Particularly, disadvantages are separated into financial and non-financial categories. Interestingly, attitudes about benefit-concern perceptions did not show any significant contribution to the model.
Model Results. An SVM approach was applied to investigate ridesourcing adoption behavior considering mode dependency and attitudinal factors. By considering balanced class-weights for a linear kernel SVM, a hyperparameter C = 1 was detected. The model performance results are illustrated in Table 18. The precision measure represents the ratio of true positives over the sum of true positives and false positives. The recall measure refers to the ratio of true positives over the sum of true positives and false negatives. Accuracy measures the ratio of all correctly predicted observations (true positives +true negatives) over the whole sample. A summary of the model performance is demonstrated in Table 18. Table 18 Linear SVM Model Performances Linear SVM Hyperparameter C 1 0.25 Class weights W-non-frequent balanced 2.5 W-frequent balanced 8 Train Data Overall Accuracy 86.40% 92.30% Precision Recall Precision Recall Majority-Non-frequent 0.99 0.86 0.98 0.94 Minority-frequent 0.4 0.92 0.56 0.78 Test data Overall Accuracy 85% 90.40% Precision Recall Precision Recall Majority-Non-frequent 0.97 0.86 0.97 0.93 Minority-frequent 0.36 0.76 0.51 0.71 The model showed an overall accuracy of 86.4% and 85% on training and test sets, respectively, with no signs of overfitting (recall value close to 1). However, a further look into the confusion matrix, reveals that the model’s accuracy is mainly due to its performance on the majority class (non-regular riders). On the contrary, the precision of minority class predictions is quite low, 0.4 and 0.36 for training and test sets, respectively. This is a critical issue given the nature of the study. The main objective of the model is to capture frequent riders, which is less than 10% of the sample. In this regard, underestimating the number of frequent riders (i.e., false negatives, or type II error) might not be as crucial as overestimating (false positives or type I error). Hence, it sounds reasonable to slightly sacrifice the recall of minority class in exchange for an increase in precision. In this regard, we further manipulated the class-weights and re-ran the grid search algorithm in search for better models. Consequently, we were able to optimize the model by increasing the misclassification penalty on the minority group and decreasing it in the majority class. In terms of contributing factors, Table 19 presents the model coefficients. It reveals that millennials showed the highest positive impact on frequent usage of ridesourcing. This...
Model Results. The model was applied to the various design storm events in the current drainage configuration pattern and is identified in the report as preconditions (Pre). Once the Preconditions model results were obtained and verified, modifications were made to the model that would simulate the proposed improvements for re-hydration of the project area. The proposed improvement model runs are identified in the report as Postconditions (Post). WilsonMiller, in order to determine the required surface water elevation, reviewed the historic flow patterns and re-establish the desirable plant communities within the project area. The surface water elevations for the desirable plant communities were established throughout the project area by the project ecologists. Peak stages for the four design events are shown for the selected basin nodes in pre and postconditions in Table 4-1. The locations of the selected basin nodes referenced in the table are shown in Exhibit 3-4. Postcondition for the basin nodes illustrate resulting peak water levels greater than or equal to the Precondition level. The larger storm events that occur less frequently reflect a slight difference in water elevations. The smaller more frequent storm events show greater increases between 0.4 and 0 feet. The model results also illustrated that after a storm had occurred in the preconditions state the peak stage would recover in less than 48 hours. Postcondition results indicate that the peak level, although equal to or slightly higher than preconditions, would remain staged for longer periods of time (240 hours and greater). Table 4-1. Peak Stages (feet, NGVD) in Breakfast Point Area for Various Design Storms DESIGN STORM EVENT CONDITION BASIN NODE N-10 N-20 N-30 N-40 N-50 N-60 100-YEAR 24-HOUR PRE 3.8 4.3 4.7 4.3 2.7 4.0 POST 4.2 4.3 4.7 4.4 2.7 4.0 25-YEAR 24-HOUR PRE 3.5 3.9 4.2 4.0 2.4 3.7 POST 3.7 4.0 4.2 4.0 2.4 3.7 10-YEAR 24-HOUR PRE 3.3 3.7 3.9 3.7 2.2 3.6 POST 3.4 3.8 3.9 3.8 2.2 3.6 5-YEAR 24-HOUR PRE 3.1 3.5 3.7 3.5 2.0 3.4 POST 3.2 3.6 3.7 3.6 2.0 3.4 DESIGN STORM CONDITION BASIN NODE N-70 N-80 N-90 N-100 N-110 N-120 100-YEAR 24-HOUR PRE 2.9 3.9 4.1 2.7 3.5 1.5 POST 2.9 3.9 4.1 2.7 3.5 1.5 25-YEAR 24-HOUR PRE 2.6 3.6 3.8 2.5 3.2 1.5 POST 2.6 3.6 3.8 2.5 3.2 1.5 10-YEAR 24-HOUR PRE 2.5 3.4 3.6 2.3 3.0 1.5 POST 2.5 3.4 3.6 2.3 3.0 1.5 5-YEAR 24-HOUR PRE 2.3 3.3 3.5 2.2 2.9 1.5 POST 2.3 3.3 3.5 2.2 2.9 1.5 DESIGN STORM CONDITION BASIN NODE N-130 N-140 N-150 100-YEAR 24-HOUR PRE 1.0 1.8 1.0 POST 1...
Model Results. Model output is used to support development of a baseline conditions and evaluation of future states under various management scenarios. Key components for interpretation of the model output include: ! Pollutant loading (annual, individual storm event) by subwatershed ! Imperviousness by subwatershed ! Flow frequency for tributaries ! In-stream water quality (annual, seasonal, and critical low flow) ! Xxxxxx Reservoir water quality (mean summer chlorophyll-a) Model results will also be used to develop a Site Assessment System (SAS) to support evaluation of the development impacts on a small scale. The system will be used to evaluate management controls to determine if these measures are sufficient to maintain pre-development conditions.
Model Results. For the baseline conditions, the aim is to verify that the model gives a broadly realistic view of the current water situation of the Kairouan aquifer. Figure 6 shows the time series for the total volume of water being input to and output from the Kairouan aquifer over the 36 month baseline simulation period. Water input to Kairouan Water output from Kairouan Water volume in Kairouan aquifer 70000 60000 Volume of water (x103 m3) 50000 40000 30000 20000 10000 0 Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Month
Model Results. Before the results are presented, the scenarios modelled will be defined. For the baseline scenario, the most up-to-date data are used that were available. Monthly recorded rainfall from within the study area for the years 1995-1997 was used, the latest years of data available. For canal inflows, data for eight local water supply canals at monthly resolution for the years 2008-2010 were used. For domestic and industrial use, and for treated wastewater reuse, only a single annual value for 2009 was available. In the absence of better data, it was assumed that these demands were evenly distributed through the year, and the annual cycle was repeated three times. For agriculture, data were available for nine crops in the study area: alfalfa, cotton, maize, rice, tree crops, summer vegetables, winter vegetables, wheat and 'other'. For each crop, the data available were: crop planted area, monthly crop water requirement per unit area, crop productivity (yield) per unit area and crop economic value per unit area. Two 2050 scenarios were simulated: a best case and a worst case, both defined with ECRI. The best case scenario, on top of decreasing rainfall totals, assumes a 10% increase to canal inflows (a proxy for changes to Nile flows), significant increases in treated waste-water volumes, and mitigation measures put in place to prevent inundation due to sea-level rise. The worst case scenario assumes decreases to canal flows, that sea- level rise results in a 13% loss to agricultural land (assumption from ECRI), that water consumption increases generally and that treated waste-water volumes do not improve on the present day. Both scenarios assume increases in domestic and industrial demand, thereby accounting for population increase and improving living standards. There are two main areas of considerable uncertainty with respect to the 2050 scenarios: a) the proportion of land that could be lost to sea-level rise. Because of the uncertainty in the actual level of sea-level rise predicted by global climate models, and because it is not known how much effort will be put into local mitigation efforts, values between 2-20% land loss were chosen to represent various levels of sea level rise and/or investment in mitigation measures. b) the direction and magnitude of changes in Nile flows. The results of numerous modelling studies have not converged to agree on either the direction or magnitude of change to Nile flows, with the current range of predictions from -50 to ...
Model Results. Under baseline conditions, model results suggest significant water surplus in the Xxxx Xxxxxxxx Xxxxx (Figure 23). Supply outstrips demand by over three times. This is not what was expected. It was expected that the water availability would be stable if not declining in this region. The most likely explanations are that a) too much water is being removed from one or more of the sources feeding the Xxxx Xxxxxxxx Canal or that b) the demands are being underestimated. Figure 23: Baseline model simulation for the Jordan River case study. Figure 23 shows an monthly water surplus of 14 000 m3 according to the input data provided by NCARE, the Jordanian WASSERMed partner. Under future conditions, it is expected that the baseline situation will deteriorate. Lower rainfall totals, reduced streamflow volumes, increasing population, improving lifestyles and increased evapotranspiration from crops suggest that while the supply is set to decrease, the demand will rise. Therefore, the 2050 simulation is expected to show a more pessimistic scenario than the baseline. Figure 24 shows the results for the 2050 simulation. Contrary to expectations, the situation improves. While the demand does indeed increase, the supply increases by a greater amount. This suggests that either the input data are over optimistic, or that the various supply sources are set to be even more overexploited than at present (Hadadin et al., 2010). Thus, this picture can be thought of as a 'false positive', as will be discussed in Section 3.3.3. Figure 24: 2050 model results for the Jordan River case study.
Model Results. This section presents the results of the Syros water balance model. Results are presented for the baseline conditions (2010) and for future scenarios (2011-2050).
Model Results. Estimation results of the proposed state-space model is shown in Table 7. In the following subsections I will discuss players’ play and purchase decisions with the presence of in-game dynamic incentive structure.
