a garbage truck could be a good option for V2G integration. Larger capacity models should have a sufficient SoC in the peak period during the summer months, and flexible routes/schedules can be integrated as well. Ultimately, our research indicates that the factor for electric garbage truck adoption will result in savings for fleet operators. Electric truck models are currently estimated to save US$35,000 per year in operating costs, while partnerships with utilities for V2G integration could result in further operational savings. Ultimately, we estimate electric garbage trucks to be a tested and available alternative by 2025–2030.
5.Individually Owned EVs
Individually owned vehicles are the most widely adopted electric vehicles in the United States. However, these vehicles as well as their chargers are not currently enabled for V2G participation there. Fortunately, a trial V2G program that is a partnership between Enel and Nissan of Europe has been underway in Denmark since 2015 [18], which allows owners of Nissan Leaf models to supply energy to the grid. If this program is successful, a similar partnership is scheduled to begin in Italy and Germany [19].
For 2018, the new Nissan Leaf model has been approved for V2G integration. It is the first EV to gain such an approval [20]. Based on our assumption that bi-directional charging equipment will become widely available in the near-future to interested EV owners in the United States, we believe that individual EVs will be a good alternative for peak power V2G integration between 2020 and 2025. Furthermore, with battery capacities increasing with each EV announced, individual EVs should have a sufficient SoC to support peak demand integration during the peak hours of 4 to 9 pm.
6.Military Fleets
The final alternative or candidate for V2G integration we selected for analysis were military non-combat vehicles. We decided to restrict military EVs to only non-combat or nontactical vehicles, since combat vehicles need to keep their SoCs as high as possible for operational readiness. We do acknowledge that combat vehicles could likely support ancillary services while plugged in, but that scenario would require a separate analysis outside our scope.
In 2013, the US Department of Defense (DoD) acquired 500 alternative fueled vehicles [21]. It is predicted that the DoD will own or lease 92,000 hybrid and electric vehicles through 2020 to help lower its fuel consumption, and reduce the risk and associated impact of fuel price volatility [22]. Furthermore, through Phase 2 of the Smart Power Infrastructure Demonstration for Energy Reliability and Security (SPIDERS) program, the DoD along with the Department of Energy and the US Army ran the first V2G test at Fort Carson, Colorado. The test integrated a 1 MW solar microgrid with five electric vehicles coupled “with advanced bi-directional vehicle chargers to integrate the battery capacity of electric vehicles in both microgrid and normal operations” [23].
With the current DoD experience in mind, we believe that future non-tactical EVs could be used during summer peak times. Such vehicles, if coupled with PV installations already in place on many military installations should also have sufficient SoCs during the 4 to 9 pm window selected for study.
6.1.Model building
At the early stage of the technology development, this study adopted the Hierarchical Decision Model (HDM), which was created and developed by Dundar Kocaoglu and Tugrul Daim [24] to better understand and track decision. The HDM is a methodology to analyze and evaluate best fitting alternatives in order to accomplish a specific objective. It uses a multicriterion that flows into alternatives selection process. Our study applied the HDM into four levels, which were Objective, Perspectives, Criteria and Alternatives, that contributed to the best option for the main objective. The HDM used the judgment of experts to prioritize the important perspectives, criteria and alternatives through the pairwise comparison technique [24, 25]. Those perspectives and criteria were then weighed by these experts, who then evaluated and estimated the complex and complicated system to gain the best decision strategically. However, the results from the HDM provided inconsistent and disagreement ratios, which indicated how much their responses did not agree with each other.
As discussed above, the objective was to determine the best opportunity behind the meter transportation technologies to use for future summer peak V2G programs. As Figure 1 shows, the model was created based on the HDM analysis to accomplish the goal. The decision model is illustrated in Figure 1. This model was created through the HDM link website to collect data from the experts. The respondents did pairwise comparison through the link for all three perspectives and a separate comparison in each node among the criteria is seen in Figure 1.
Figure 1.The HDM in four levels.
Finally, the experts completed weighing the pairwise comparison of all the perspectives, criteria and potential alternatives. Then, their opinion is submitted to the model and contributed to the result as the best opportunity of technology options.
6.2.Data analysis and results
The HDM results showed the best option of potential alternatives through the highest score. Moreover, there was a critical statistic result which is the inconsistency that explores how consistent and careful the experts weighted different factors. The standard acceptable rate for inconsistency was less than 0.1. If it had been more than 0.1, the quality of judgment should not be considered [26]. However, it also depends on the variety of perspectives, criteria and different tolerance levels. In this study, the inconsistency for each expert was less than 0.1, as shown in Table 2, therefore the results from all the experts can be considered as consistent judgment. Moreover, it shows that the disagreement rate is less than 0.1, which means that all the experts are in the same agreement with regard to weighting the criteria and perspectives relating to the objective.
The F-test value was calculated through pairwise comparison in the HDM model from all the participating experts, as shown in Table 3. The value indicated a degree of agreement due to the benchmark value of 2.33 at 0.1 level (90% confidence level) and the final value of 2.61, which is over 2.33. Therefore, it proves that the HDM weights from the selected experts were in agreement with a 90% confidence.
Table 2.HDM results based on the alternatives.
Table 3. HDM statistical results.
Through the pairwise comparison in HDM methodology, the important perspectives and criteria reveal overall
