Predicting Fuel Consumption by Artificial Neural Network (ANN) Based on the Regular City Bus Lines
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsDear Author,
In your manuscript you provided the analysis of fuel consumption as a result of forecasting fuel consumption. There are expected elements of the scientific manuscript, but there are, also, issues which need to be clarified: e.g., you mention "the factors influencing electric vehicle adoption" (Page 2). It is uncleary how is this is related to the topic of the manuscript.
In sectioon 3.1 a huge number of different factors to be considered is mentioned. Are all of them are involved in your model (which is left unclear by which formula, or AI detailed approach is made).
Comments on the Quality of English Language
The quality of English language is generally good, but a help of native English speaker might improve it.
Author Response
I sincerely thank the reviewer for their insightful comments and constructive feedback, which have significantly improved the quality of our manuscript. Your suggestions have been invaluable in refining our work. I appreciate your time and effort in reviewing our paper.
In your manuscript you provided the analysis of fuel consumption as a result of forecasting fuel consumption. There are expected elements of the scientific manuscript, but there are, also, issues which need to be clarified: e.g., you mention "the factors influencing electric vehicle adoption" (Page 2). It is uncleary how is this is related to the topic of the manuscript.
Answer: I deleted mentions about electric vehicles. The paper focuses on the fuel vehicles, so it is not related.
In sectioon 3.1 a huge number of different factors to be considered is mentioned. Are all of them are involved in your model (which is left unclear by which formula, or AI detailed approach is made).
Answer: The presented factors show the previous research and focus on the problem of developing prediction models based on traditional algorithms. Presented factors play a crucial role in fuel consumption, so they should be taken into account in the prediction model. Because of plenty of factors, the best possibility to develop a prediction model is using AI tools. In the conducted research the ANN models were examined and developed in the Matlab 2024b.
I added the description of made research and comparison of examined models – subchapter 3.2. and 3.3.
Reviewer 2 Report
Comments and Suggestions for Authors- Can the authors provide more specific quantitative data or comparative analysis to highlight the environmental and economic benefits of the measures discussed (e.g., free parking and tax exemptions for BEVs)?
- Were any uncertainties or limitations identified in the cited studies (e.g., differences in pollutant emissions among regions or vehicle types)? If so, how were these addressed?
- The methodology for fuel consumption forecasting is detailed. How were the correction coefficients for vehicle and route-specific factors derived and validated?
- The thermal stabilization effects seem to vary significantly with ambient temperature. Did the authors include regional climate variations in their modeling?
- Recently, very important studies have been carried out on fuel conception and Forecasting. Please mention these 3 studies in detail to increase the depth of the article.
Converting lignocellulosic biomass into valuable end products for decentralized energy solutions: A comprehensive overview
https://doi.org/10.1016/j.seta.2024.104065
Improving Productivity at a Marble Processing Plant Through Energy and Exergy Analysis
https://doi.org/10.3390/su162411233
Development of Comparative Forecasting Models of Daily Prices of Aggressive Pension Mutual Funds by Univariate Time Series Methods
https://doi.org/10.34110/forecasting.1465436
- The developed software system includes modules for route, vehicle, driver, and trip management. Can the authors provide examples or case studies demonstrating the system's practical application?
- In Figure 6, the prediction error is presented. Could the authors explain how this error was minimized during model development?
- The authors highlight the benefits of the system for reducing operational costs. Can they quantify these benefits based on their findings (e.g., percentage cost savings)?
Author Response
I sincerely thank the reviewer for their insightful comments and constructive feedback, which have significantly improved the quality of our manuscript. Your suggestions have been invaluable in refining our work. I appreciate your time and effort in reviewing our paper.
- Can the authors provide more specific quantitative data or comparative analysis to highlight the environmental and economic benefits of the measures discussed (e.g., free parking and tax exemptions for BEVs)?
Answer: The state of the art was corrected, and the paper not directly connected with the research area was deleted, i.e. BEVs free parking and tax. I apologize for the inconvenience.
- Were any uncertainties or limitations identified in the cited studies (e.g., differences in pollutant emissions among regions or vehicle types)? If so, how were these addressed?
Answer: Fuel consumption plays an important role in economics. The cost of fuel for transport companies is about 30-40% of total costs. What is more many cities in the city centre implement a clean transport zone, i.e. Cracow (Poland) from 2025 where the first implementation of the proposed model was done. In different countries also the clean transport zone was implemented, i.e. London (2008), Florence (1991), Sztokholm (2010), Madrid (2018), Amsterdam (2008), Oslo (2017), Brussels (2018). The terrain and location of the city play the main role in the resistance to pollution. Big cities with many roads, concrete, and high buildings are susceptible to pollution due to limited air movement. If there is also an airport near the cities, problems cumulate pollution from road traffic and air traffic. Also, the type of fuels and vehicles' technical condition has great impacts – diesel engines are worse than petrol or CNG, so producers must declare the pollution emissions by EURO standards.
- The methodology for fuel consumption forecasting is detailed. How were the correction coefficients for vehicle and route-specific factors derived and validated?
Answer: Presented details for forecasting could be implemented just in the laboratory – where the engine and vehicle are monitored by many sensors. In typical companies detailed telematics and monitoring the movement on the route are not possible – the GPS locations are not enough. Because of this in the research, the general information about the route, fuel consumption and vehicle was used as input data. In the company where the firsts implementation and tests the presented details for each course were collected: Route length [km], number of bus stops, probability of traffic jams [from 1-low to 3-high], ambient temperature [°C], external database, technical state of the vehicle [from 1-good to 5-bad], type of petrol (1 - ON; 2 - E95), filling of the vehicle/number of passengers (from 1- empty to 5-full). Based on this data the presented model was developed.
I added the description of the input dataset and model used ANN.
- The thermal stabilization effects seem to vary significantly with ambient temperature. Did the authors include regional climate variations in their modeling?
Answer: Yes, the ANN model used the ambient temperature from the Polish weather research centre (https://imgw.pl/).
- Recently, very important studies have been carried out on fuel conception and Forecasting. Please mention these 3 studies in detail to increase the depth of the article.
Converting lignocellulosic biomass into valuable end products for decentralized energy solutions: A comprehensive overview https://doi.org/10.1016/j.seta.2024.104065
Improving Productivity at a Marble Processing Plant Through Energy and Exergy Analysis https://doi.org/10.3390/su162411233
Development of Comparative Forecasting Models of Daily Prices of Aggressive Pension Mutual Funds by Univariate Time Series Methods https://doi.org/10.34110/forecasting.1465436
Answer: I read the mentioned paper and cited two of them, connected with the paper.
- The developed software system includes modules for route, vehicle, driver, and trip management. Can the authors provide examples or case studies demonstrating the system's practical application?
Answer: The model was just developed a few months ago, and implemented as prototype software. The company just started using this, so the results of using the system can not be presented at this moment. However, I added the analysis of performance for the implementation in the software system ANN model.
- In Figure 6, the prediction error is presented. Could the authors explain how this error was minimized during model development?
Answer: This error was not minimalized. The error presented in the figure shows the difference between predicted fuel consumption the real fuel consumption. The minimalization of error was done in the developing model stage. I added subchapter 3.2. and 3.3. to the paper to present the main research related to developing this model. In this description, the minimalization of error (MSE) was presented.
- The authors highlight the benefits of the system for reducing operational costs. Can they quantify these benefits based on their findings (e.g., percentage cost savings)?
Answer: Before finishing the software tests it is not possible, Based on the actual state we expect about 3-6% of savings.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe abstract could be clearer by explicitly stating the specific contributions of the study. For example, how the proposed system uniquely improves real-time forecasting compared to existing methods. The following related research can be compared: a) Short-term Load Forecasting of Distribution Transformer Supply Zones Based on Federated Model-Agnostic Meta Learning b) Multi-node load forecasting based on multi-task learning with modal feature extraction
In Figure 1, the explanation of fuel consumption during idle states at various temperatures lacks clarity on how these values were obtained. Were these empirical measurements, or derived from simulation models?
Table 1 is could add a brief explanation about how boundary conditions were simulated or validated against real-world scenarios.
The explanation of the technical condition of vehicles in Section 3.1.6 is thorough. However, the claim regarding spark plug replacement (every 15,000–20,000 km) may vary across vehicle manufacturers. Please give more details
Figure 5 should be improved largely
The conclusion should add specific performance metrics. For instance, how much fuel consumption was reduced on average or provide examples of cost savings achieved during system implementation.
Author Response
I sincerely thank the reviewer for their insightful comments and constructive feedback, which have significantly improved the quality of our manuscript. Your suggestions have been invaluable in refining our work. I appreciate your time and effort in reviewing our paper.
The abstract could be clearer by explicitly stating the specific contributions of the study. For example, how the proposed system uniquely improves real-time forecasting compared to existing methods. The following related research can be compared: a) Short-term Load Forecasting of Distribution Transformer Supply Zones Based on Federated Model-Agnostic Meta Learning b) Multi-node load forecasting based on multi-task learning with modal feature extraction
Answer: I read and analysed the proposed research and mentioned it in the paper.
In Figure 1, the explanation of fuel consumption during idle states at various temperatures lacks clarity on how these values were obtained. Were these empirical measurements, or derived from simulation models?
Answer: These results presented in Figure 1 were obtained by empirical measurements.
Table 1 is could add a brief explanation about how boundary conditions were simulated or validated against real-world scenarios.
Answer: Table 2 presents the research from different papers and authors. The results presented in the paper were obtained from laboratory road research.
The explanation of the technical condition of vehicles in Section 3.1.6 is thorough. However, the claim regarding spark plug replacement (every 15,000–20,000 km) may vary across vehicle manufacturers. Please give more details.
Answer: Yes it is. Of course, it is related to the exact type and model of the vehicle. The mentioned 15k-20k km is recommended by Polish car services. In added ANN model presented in subchapter 3.2. and 3.3. this was generally related to the Technical state of the vehicle.
Figure 5 should be improved largely
Answer: I change this figure. The figure presents the screen from the implemented system (it is a prototype version), so I changed the legend place and readability of the figure.
The conclusion should add specific performance metrics. For instance, how much fuel consumption was reduced on average or provide examples of cost savings achieved during system implementation.
Answer: Before finishing the software tests it is not possible. Based on the actual state we expect about 3-6% of savings. I added subchapter 3.2. and 3.3. to the paper to present the main research related to developing this model. In this description, the minimalization of error (MSE) was presented.
Reviewer 4 Report
Comments and Suggestions for Authors- The paper should clearly highlight the major contribution of the model. The proposed methodology appears to primarily compile existing knowledge and practices, which is lack of novelty.
- I suggest the author revise the title, abstract, and keywords of the paper, they don't reflect the content and focus of the main body.
- The paper doesn't define key parameters, main assumptions. And it lacks sufficient details for experiment setup.
- The paper lacks experimentation and validation result. For example, it doesn't address how to calibrate the parameters, account for different types of vehicles, or consider external conditions.
- I recommend that the author incorporate recent work on fuel consumption in Section 2. For instance the recent work "Neural network surrogate models for aerodynamic analysis in truck platoons: Implications on autonomous freight delivery"
- The writing needs significant improvement. The paper is poorly organized, contains redundancies and grammar errors.
Author Response
I sincerely thank the reviewer for their insightful comments and constructive feedback, which have significantly improved the quality of our manuscript. Your suggestions have been invaluable in refining our work. I appreciate your time and effort in reviewing our paper.
- The paper should clearly highlight the major contribution of the model. The proposed methodology appears to primarily compile existing knowledge and practices, which is lack of novelty.
Answer: The presented factors show the previous research and focus on the problem of developing a prediction model based on traditional algorithms. Presented factors play a crucial role in fuel consumption, so they should be taken into the account in the prediction model. Because of plenty of factors, the best possibility to develop a prediction model is using AI tools. In the conducted research the ANN models were examined and developed in the Matlab 2024b.
I added the description of made research and comparison of examined models – subchapter 3.2. and 3.3.
- I suggest the author revise the title, abstract, and keywords of the paper, they don't reflect the content and focus of the main body.
Answer: I revised and changed the title, abstract, and keywords of the paper.
- The paper doesn't define key parameters, main assumptions. And it lacks sufficient details for experiment setup.
Answer: I added the description of made research and comparison of examined models – subchapter 3.2. and 3.3.
- The paper lacks experimentation and validation result. For example, it doesn't address how to calibrate the parameters, account for different types of vehicles, or consider external conditions.
Answer: I added the description of made research and comparison of examined models – subchapter 3.2. and 3.3.
- I recommend that the author incorporate recent work on fuel consumption in Section 2. For instance the recent work "Neural network surrogate models for aerodynamic analysis in truck platoons: Implications on autonomous freight delivery
Answer: Thank you for suggesting nice research related to the presented works. I read this and mentioned it in the paper.
Round 2
Reviewer 4 Report
Comments and Suggestions for Authors Thanks for the revision. I have some minor comments:- The quality and the resolution of the figures should be improved.
- the practical implication should be discussed.
Author Response
I sincerely thank the reviewer for their insightful comments and constructive feedback, which have significantly improved the quality of our manuscript. Your suggestions have been invaluable in refining our work. I appreciate your time and effort in reviewing our paper.
I changed the paper regarding your comments.
