Integrated Production–Logistics Scheduling in Flexible Assembly Shops Using an Improved Genetic Algorithm

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is an interesting article on production-logistics scheduling in flexible assessing an improved genetic algorithm.

The following comments should be considered for improving the manuscript:

• Line 41: How traditional assembly lines have poor performance.
• Kine 45: Is responsiveness the problem? Authors needs to clarify which industry is in focus.
• Line 57: How production and logistics are integrated. How about inventory?
• Line 61: Provide a source for two decades?
• Line 85: Explain de-coupling as the client intervention comes very early in the supply chain.
• Section 2.3: Provide the relevant studies from which challenges were extracted. How critical are these challenges? What are other potential solutions to address the challenges, and why is the proposed solution suitable?
• Table 4: How parameters were established.
• Which software was used for algorithm simulations
• Figure 3: Why 10% random path blockage?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper needs revision. The main questions that arise are as follows:

1. The method used for comparison (two-stage heuristic method) needs citations. It is not clear what this method is.

2. In line 491, F is defined as F = 1/Cmax. In line 539 as F = Cmax (Eq. 11). In line 613, again as F = 1/Cmax. Please pay attention.

3. Table 4: 

3.1. How are the values of 0.45 and 0,0075 selected? Are these the optimal values? Why is the maximum value of crossover probability 0.6?

3.2. Please, provide the expressions for the adaptive crossover and mutation probabilities. Are they dynamically adjusted based on formula/expression?

4. It is not clear how many sets of 20 independent runs are performed for scenario 1?

4.1. Why are only 200 iterations performed? From Fig. 5 it is clear that the algorithms are still improving the value of the objective function. With another 50 or 100 iterations, there will be better results.

4.2. The results in Fig. 4 do not correspond to the result in Fig. 5. If the results are from one set, there is something wrong. The lowest value of IGA is under 80, according Fig. 5, but according Fig. 4, the value of 80 is not reached.

4.3. The results in Table 5 and in Fig. 5 do not correspond. In Fig. 5 the best value obtained by IGA is smaller than 80; in Table 5, the listed value is 82.1. The GA has the best result of around 83 (outlier); the next best value is around 88. In Table 5. the best value of GA is 89.6.

4.4. In lines 844-847 the results for the two-stage and GA are discussed. Why is the conclusion for two-stage and IGA?

5. Conclusions

5.1. Which results show that "the adaptive crossover and mutation operators dynamically balance exploration and exploitation during evolution. The proposed mechanism effectively prevents premature convergence, thereby enhancing overall search efficiency and population diversity." And, again, how exactly is the dynamic adaptation performed?

Comments on the Quality of English Language

One more English review will only enhance the quality of the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have taken into account all comments and remarks and have made the necessary changes and additions to the text of the paper.