The Influence of Public Transport Delays on Mobility on Demand Services

Round 1

Reviewer 1 Report

authors have presented, descriptive analysis of the public transit data set yields that delays and MoD demand both highly dependent on the weekday and time of day as well as the location within the city, and that delays in the city and in consecutive time intervals are correlated.
My comments are as follows:

• the manuscript is written well. English writing is good, need to check some typo errors to look into.
• explain figure 3 more about delay propogation.
• add some more recent work on the transportation system in the introduction and related work as suggested below:

M. Waqas, S. Tu, S. U. Rehman, Z. Halim, S. Anwar et al., “Authentication of vehicles and road side units in intelligent transportation system,” Computers, Materials & Continua, vol. 64, no. 1, pp. 359–371, 2020.

J. Liu, X. Kang, C. Dong and F. Zhang, “Simulation of real‐time path planning for large‐scale transportation network using parallel computation,” Intelligent Automation & Soft Computing, vol. 25, no.1, pp. 65–77, 2019.

B. Yan, “Improvement of the economic management system based on the publicity of railway transportation products,” Intelligent Automation & Soft Computing, vol. 26, no.3, pp. 539–547, 2020.

Singh, S., Sharma, P.K., Yoon, B., Shojafar, M., Cho, G.H. and Ra, I.H., 2020. Convergence of blockchain and artificial intelligence in IoT network for the sustainable smart city. Sustainable Cities and Society, 63, p.102364.

H. Gao, W. Huang and X. Yang, “Applying probabilistic model checking to path planning in an intelligent transportation system using mobility trajectories and their statistical data,” Intelligent Automation & Soft Computing, vol. 25, no.3, pp. 547–559, 2019.

W. Liu, Y. Tang, F. Yang, C. Zhang and D. Cao, “Internet of things based solutions for transport network vulnerability assessment in intelligent transportation systems,” Computers, Materials & Continua, vol. 65, no. 3, pp. 2511–2527, 2020.

Author Response

Thank you for your comments.To explain figure 3 better, we added the following sentence on page 5 and 6: "It shows the probability that an above average delay at Marienplatz (left) or Münchner Freiheit (right) correlates with an above average delays at other stations. Yellow shading refers to a high covariance, blue shading to a low covariance. Clearly, stations along the same line have a higher probability of delays, and at Marienplatz, this mainly affects the East-West connection (S-Bahn), not so much the North-South connection (U-Bahn). The covariance between the selected station and stations along other lines is low, and the remaining covariance can be due to external influences, customers transferring lines, or intersecting lines. From the strong spatial and temporal differences in delays, we conclude that our spatial and temporal resolution is reasonable."

We believe that the suggested papers do not strengthen our paper. In particular:

• Authentication of Vehicles and Road Side Units in Intelligent Transportation System:
  Has no significant overlap to our research objectives. We don't focus on automated vehicles, rather than security threats.
• Simulation of Real‐Time Path Planning for Large‐Scale Transportation Network Using Parallel Computation:
  Has no significant overlap to our research objectives. We don't use traffic simulations in our approach, as this is a data analyis on real world observed data.
• Improvement of the Economic Management System Based on the Publicity of Railway Transportation Products:
  Again, we cannot see any added value from this paper. We don't focus on economic management systems of public railways systems.
• Convergence of block chain and artificial intelligence in IoT network for the sustainable smart city:
  Has no overlap to our research objectives: How could block chain help to reduce delays in public transport networks, or to analyse it's effect on mod systems?
• Applying Probabilistic Model Checking to Path Planning in an Intelligent Transportation System Using Mobility Trajectories and Their Statistical Data:
  Interesting, paper we might include this in a further study, where we focus on the effect of travel time on peoples mode choice behavior. For this analysis we don't see an added value.
• Internet of Things Based Solutions for Transport Network Vulnerability Assessment in Intelligent Transportation Systems:
  Has no significant overlap to our research objectives. As mentioned for the last paper, we might take this into account in a further study on mode choice behavior.

Reviewer 2 Report

The authors investigate the influence of public transport delays on moility on demand services, also designing a descriptive analysis.

The proposed study is interesting but there are some points that the authors should better discuss.

The authors should be better described the novelties of their study with respect to existing ones. In particular, the author should discuss limitation and cons of the examined approaches. Furthermore, the authors should provide more details and discussion about the obtained results. The Discussion section also needs to be improved by analyzing the outcome of evaluation section.

I suggest to further analyze more recent approaches about the examined topics. In particular, I suggest the following papers to further investigate graph-based machine learning and emotional mood approaches for COVID pandemic in the introduction section:

1) An Epidemiological Neural network exploiting Dynamic Graph Structured Data applied to the COVID-19 outbreak. IEEE Transactions on Big Data.

2) An emotional recommender system for music. IEEE Intelligent Systems.

Finally, I suggest to perform a linguistic revision.

Author Response

Thank you for your review.

To distinguish this paper from existing literature, we improved the exposition of the literature review by adding three statements. They show why we cannot use existing approaches for demand prediction and impact analysis, and how our research differs from existing impact analysis literature which is mainly driven by the surveys or exemplary data analysis.

We improved the introduction by also referring to the challenges and opportunities of conducting this study during the COVID period.

We also added a paragraph in the discussion and conclusion on page 14 outlining (i) that our results are low values (which is to be expected), and (ii) how our findings could be strengthened with more data. "}Further, we must mention that the effect is little. This is because the S-Bahn has very similar and rather low delays on most instances. Since the data set only permits integer delays and since delays are subject to external influences, effects remain little. More precise data would permit a more extensive analysis. However, this approach is important since it allows third parties such as policy makers or new market entrants, to measure the effect."

We added one of the suggested papers (La Gatta et al., 2020) in our exploratory data analysis on COVID.

We did not add the other paper, because we don't see any relationship to our research objective. 

Round 2

Reviewer 2 Report

I think that the authors have addressed all my concerns.