Round 1

Reviewer 1 Report

General comments

The paper ‘A novel two-phase approach to forest harvesting optimization using cable logging’ is mostly easy to read and follow, and the method, results and conclusions are all well aligned. I have some specific questions and suggestions to improve the paper. Me being closer to practice I suggest that you enhance that connection, underlying what kind of time or money potential this type of work can have and, if possible, compare those figures to your real world scenario. Moreover, it would also be interesting to compare a real world planning outcome to your suggested mathematical approach.

Language

Just a few spelling errors noted and some sentences that could be adjusted.

Specific comments

Abstract

The term “terrestrial timber harvesting” is new for me (although I can understand what you are going for), is there another term that is synonymous?

Introduction

It would also be good to mention that the timber value is important when choosing method, and if the value are too low, extraction with logging towers or some aerial methods can at some point not even be justified.

Do you have any references that could support the extraction process described in Fig. 1? It would be interesting since you, for example, claim that “recent pilot and demonstrative experiences have reported using unmanned aerial vehicles (UAV) to enhance worker safety”

Related Work and Problem Description

Regarding the tree extraction process on which you base this work: As I understand this, you are constrained by some prerequisites. One important one seems to be the initial positions of the yarders. Another possibly important aspect is the ground topology. Has that been considered in any way? Why/why not? Could you please describe your initial constraints and assumptions a bit further?

In your constraints descriptions, what is the definition of M and k?

In Fig. 5, isn’t it five towers in total?

Fig 5: I do not understand if this is an actual scenario that has happened. This is also connected to Figure 13; how do you explain going from the “Deep Learning Engine” step to the “Optimization Problem”? I think the plotting need to be explained further (as it is not a part of [15] either).

Figures 14 – 16 could benefit from having the names written out in the figures or in the captions.

In your conclusions it would be interesting if you had assessed what this could mean for the enterprises using this kind of solution. Do you have any outlook of the future possibilities for implementation?

Just a few spelling errors noted and some sentences that could be adjusted.

Author Response

General comments

The paper ‘A novel two-phase approach to forest harvesting optimization using cable logging’ is mostly easy to read and follow, and the method, results and conclusions are all well aligned. I have some specific questions and suggestions to improve the paper. Me being closer to practice I suggest that you enhance that connection, underlying what kind of time or money potential this type of work can have and, if possible, compare those figures to your real-world scenario. Moreover, it would also be interesting to compare a real-world planning outcome to your suggested mathematical approach.

A:  Thank you for your comment. We made changes to emphasize the quantitative comparison made with real-world scenarios (i.e., that obtained from images with Deep Learning) and also the approximate real-world planning (i.e., a greedy approach or MPA).

Language

Just a few spelling errors noted and some sentences that could be adjusted.

A: We make the language changes suggested by you. Thanks for your observation.

Specific comments

1. Abstract

The term “terrestrial timber harvesting” is new for me (although I can understand what you are going for), is there another term that is synonymous?

A: Thank you for pointing that out. We've omitted the term "terrestrial" for clearer comprehension. We hope this enhances understanding.

2. Introduction

It would also be good to mention that the timber value is important when choosing method, and if the value are too low, extraction with logging towers or some aerial methods can at some point not even be justified.

A: Thanks for your observation. We added a paragraph in the introduction section explaining this idea (economic dimension).

Do you have any references that could support the extraction process described in Fig. 1? It would be interesting since you, for example, claim that “recent pilot and demonstrative experiences have reported using unmanned aerial vehicles (UAV) to enhance worker safety”

A:  While searching the academic literature extensively, we could not identify any papers that robustly validate this specific forestry process. Nevertheless, for a practical perspective, we have referenced a webpage that delves into the innovative use of drones for the cable logging industry: https://www.saif.com/saifstoriespage/bighorn-logging.html

Recognizing the importance of precision and clarity, we've opted to omit the drone description from Figure 1, focusing instead on highlighting the expertise of individuals proficient in this technique. We appreciate your comment.

Related Work and Problem Description

Regarding the tree extraction process on which you base this work: As I understand this, you are constrained by some prerequisites. One important one seems to be the initial positions of the yarders. Another possibly important aspect is the ground topology. Has that been considered in any way? Why/why not? Could you please describe your initial constraints and assumptions a bit further?

A: Thank you for your keen observations and questions related to the ground topology process. We would like to provide further clarification on this matter:

• Planning Generation Approaches (Model and Heuristic): Our methodologies, both the model and heuristic approaches, have been designed to be adaptable to various topologies. While they possess the capability to address different terrains, we chose to focus on the current topology in our experiments because we believe it to be the most representative and relevant to the objectives of our study.
• Study Topologies: To provide a comprehensive evaluation, two distinct topologies were created for our research: one that's artificially generated and another based on a real-world scenario. Each has its unique configuration, and our findings demonstrate that our approaches can successfully address both.

In light of your feedback, we have incorporated an additional paragraph in the conclusion to further address this observation. We hope this amendment offers a more comprehensive understanding of our study's scope and assumptions.

3. In your constraints descriptions, what is the definition of M and k?

A: Thanks for your question. We add the description in our problem formulation.

4. In Fig. 5, isn’t it five towers in total?

A: Thanks for your observation. You are totally right. The correction has been made.

5. Fig 5: I do not understand if this is an actual scenario that has happened. This is also connected to Figure 13; how do you explain going from the “Deep Learning Engine” step to the “Optimization Problem”? I think the plotting need to be explained further (as it is not a part of [15] either).

A: Thank you for your observation. In response, we have enhanced the description in the paragraph corresponding to Figure 3 by incorporating an additional sentence. Furthermore, we have thoroughly restructured the paragraph detailing the acquisition of the real-world scenario, which can be found in the subsection titled "Real-world instances."

6. Figures 14 – 16 could benefit from having the names written out in the figures or in the captions.

A:  Thanks for your observation. We have applied changes to these figures. We hope they meet your requirements.

7. In your conclusions it would be interesting if you had assessed what this could mean for the enterprises using this kind of solution. Do you have any outlook of the future possibilities for implementation?

A:  Thank you for your feedback. We've added a paragraph addressing your observation. We hope it meets your expectations.

Reviewer 2 Report

Dear authors,

Cable crane yarders are most usable forest machines on steep terrains, altough cable assisted machines like forwardes are used more often in last time instead them because they are cheaper and its productivity is statisfying. But every improvement of using of any kind of the machine is welcome.

In the paper you presents an integer programming mathematical model for determining the optimal positioning of cable crane yarders conditioned to logging lines, timber logging time, logging cycle time and use a two-phase heuristic algorithm to address the problem. 

Abstract

Please write some results and final conclusion(s) in the abstract.

Introduction

In the Introduction part you have a very little citations for the statements. For example in part between line 25 and 50 concerning operation of cable yarder, and also between line 51 and 55 when you write about benefits for soil, and so on.

(possible references:

Gallo, R., Visser, R. i Mazzetto, F. (2021). Developing an Automated Monitoring System for Cable Yarding Systems. Croatian Journal of Forest Engineering, 42 (2), 213-225. https://doi.org/10.5552/crojfe.2021.768

Mologni, O., Marchi, L., Lyons, C.K., Grigolato, S., Cavalli, R. i Röser, D. (2021). Skyline Tensile Forces in Cable Logging: Field Observations vs. Software Calculations. Croatian Journal of Forest Engineering, 42 (2), 227-243. https://doi.org/10.5552/crojfe.2021.722

Knobloch C, Bont LG. A new method to compute mechanical properties of a standing skyline for cable yarding. PLoS One. 2021 Aug 19;16(8):e0256374. doi: 10.1371/journal.pone.0256374

Leo Gallus Bont, Laura Ramstein, Fritz Frutig & Janine Schweier (2022) Tensile forces and deflections on skylines of cable yarders: comparison of measurements with close-to-catenary predictions, International Journal of Forest Engineering, 33:3, 195-216, DOI: 10.1080/14942119.2022.2051159)

Graphs presented in Fig. 5, 12, 14, 15, 16, and 17 do not have axis labels and measuring units. Please change  that.

In methods you describe the calculating procedure very well, but I didn`t notice that you somewhere mentioned that all this calculation, especially time for mounting and dismounting extraction lines, an also working cycle is influenced by the terrain and weather conditions, and also by a crew organization. Productivity and extraction time is influenced by a capacity of the whole cable yarder, especially of the carriage capacity, way of the processing on the landing site, capacity of the landing site and the organization of the extraction logs from the landing site.

Also I noticed that you use terminology that is not common in forestry vocabulary, like aerial cable which is called skyline, collection site is landing site, and so on, but that is not so important in this case.

I suppose that the calculation and presented results are correct, and this approach of the work optimisation is something new which common foresters doesn`t understand too much. In Conclucions you mentioned that this is an expermental research, and maybe this article is more suitable to be published in some journal with topic related to deep learning.

Kind regards 

Author Response

General Comments

• Cable crane yarders are most usable forest machines on steep terrains, altough cable assisted machines like forwardes are used more often in last time instead them because they are cheaper and its productivity is statisfying. But every improvement of using of any kind of the machine is welcome.

In the paper you presents an integer programming mathematical model for determining the optimal positioning of cable crane yarders conditioned to logging lines, timber logging time, logging cycle time and use a two-phase heuristic algorithm to address the problem. 

A: Thank you for your comment.

Abstract

• Please write some results and final conclusion(s) in the abstract.

A: Thanks for your observation. We have added a phrase to the end of the abstract.

Introduction

• In the Introduction part you have a very little citations for the statements. For example in part between line 25 and 50 concerning operation of cable yarder, and also between line 51 and 55 when you write about benefits for soil, and so on.

(possible references:

Gallo, R., Visser, R. i Mazzetto, F. (2021). Developing an Automated Monitoring System for Cable Yarding Systems. Croatian Journal of Forest Engineering, 42 (2), 213-225. https://doi.org/10.5552/crojfe.2021.768

Mologni, O., Marchi, L., Lyons, C.K., Grigolato, S., Cavalli, R. i Röser, D. (2021). Skyline Tensile Forces in Cable Logging: Field Observations vs. Software Calculations. Croatian Journal of Forest Engineering, 42 (2), 227-243. https://doi.org/10.5552/crojfe.2021.722

Knobloch C, Bont LG. A new method to compute mechanical properties of a standing skyline for cable yarding. PLoS One. 2021 Aug 19;16(8):e0256374. doi: 10.1371/journal.pone.0256374

Leo Gallus Bont, Laura Ramstein, Fritz Frutig & Janine Schweier (2022) Tensile forces and deflections on skylines of cable yarders: comparison of measurements with close-to-catenary predictions, International Journal of Forest Engineering, 33:3, 195-216, DOI: 10.1080/14942119.2022.2051159)

A: Thank you for your suggestion. We've incorporated the references into a section that underscores the significance of the cable monitor, aligning it with insights from recent publications.

• Graphs presented in Fig. 5, 12, 14, 15, 16, and 17 do not have axis labels and measuring units. Please change that.

A: Thanks for your suggestion, we have added axis labels to figures 5, 12, 14, 15, 16, and 17.

Specific comments

• In methods you describe the calculating procedure very well, but I didn`t notice that you somewhere mentioned that all this calculation, especially time for mounting and dismounting extraction lines, an also working cycle is influenced by the terrain and weather conditions, and also by a crew organization. Productivity and extraction time is influenced by a capacity of the whole cable yarder, especially of the carriage capacity, way of the processing on the landing site, capacity of the landing site and the organization of the extraction logs from the landing site.

A: Thank you for your observation. At present, our approach operates by considering installation and harvest times, approximated to real-world durations, using extraction cables. The specifics of the problem parameters are detailed in Table 1. Additionally, we've introduced a new paragraph in the "Experimental evaluation" section. It's worth noting that our current approach emphasizes essential timings in two dimensions, as illustrated throughout the document. Incorporating a third dimension would enable us to address the comprehensive study parameters you've suggested.

• Also I noticed that you use terminology that is not common in forestry vocabulary, like aerial cable which is called skyline, collection siteis landing site, and so on, but that is not so important in this case.

A: Thanks for your observation. We have replaced aerial cable and collection site.

• I suppose that the calculation and presented results are correct, and this approach of the work optimisation is something new which common foresters doesn`t understand too much. In Conclucions you mentioned that this is an expermental research, and maybe this article is more suitable to be published in some journal with topic related to deep learning.

A:  Thank you for your valuable feedback and observations. Although the main focus of the document lies in Operational Research, we believe that this article can serve as a reference for applying Metaheuristic techniques in the planning of extraction processes, aiming to minimize time. The Forest journal is recognized as one of the leaders in the field, and the combination of theory with practical experiments is a recurring topic in many of its articles. We trust that by incorporating the modifications suggested by you, we will enrich the presented content.

Round 2

Reviewer 2 Report

Dear authors,

Thank you for the reply.

The manuscript is now much better.

Kind regards