Skip to main content
PLOS One logoLink to PLOS One
. 2026 Oct 1;21(10):e0357258. doi: 10.1371/journal.pone.0357258

Optimal parameters of roadway supports for water prevention based on the FFA algorithm

Chao Xu 1,*
Editor: Zhenhua Li2
PMCID: PMC13630240  PMID: 42821540

Abstract

Roadways are important paths in mining engineering. If the stability of roadway supports cannot be guaranteed, workers’ lives will be threatened. Therefore, the firefly algorithm (FFA) and FLAC3D numerical simulation were combined. Taking the roadway of the Banji Coal Mine as an example, the FFA was used to obtain the optimal grouting area and cable distribution design. This support scheme was then applied to engineering practice. The numerical simulation results were in good agreement with those of engineering practice, indicating that the proposed method is reliable and feasible and can be applied to determine the optimal design for supporting roadways in engineering practice.

1. Introduction

With the continuous deepening of coal mines and the expansion of the scope of roadway excavation, water inrush disasters have become among the most serious geological hazards, as they restrict safe mining, threaten the lives of underground workers and cause considerable economic losses for mining enterprises [1–5]. Roadway supports are the core measure used to control surrounding rock deformation and prevent water inrush, and the support parameters directly affect the stability of the rock surrounding roadways as well as water prevention and control. Reasonable optimal designs of roadway support parameters can not only improve the antideformation ability of the surrounding rock and reduce the occurrence probability of water inrush accidents [6–8] but also save support materials and reduce engineering costs, which has important theoretical significance and engineering application value for realizing safe, efficient and economical mining [9–13].

To determine the optimal roadway support scheme that can effectively ensure the stability of rock surrounding roadways, resist water damage, and meet economic engineering requirements, a series of systematic and in-depth experiments were conducted. These experiments focused mainly on understanding the complex fluid‒mechanical coupling effects in rock masses surrounding roadways under different geological conditions (such as different water-bearing strata, fractured rock masses, and burial depths) [14,15]. During the experimental process, various parameters, including rock mass permeability, stress distribution, deformation characteristics, and water seepage, were monitored in real time. On the basis of a comprehensive analysis of experimental data, corresponding water prevention measures and roadway support schemes were proposed from the perspective of controlling engineering costs to ensure that the proposed schemes not only have good stability and water resistance but also avoid excessive economic investment in engineering [16].

With respect to numerical simulation, which is an important auxiliary means for designing roadway supports, researchers usually propose a variety of alternative roadway support schemes first. These schemes differ in terms of the key support parameters, such as the type of support material (e.g., anchor rod, anchor cable, or shotcrete), support spacing, support strength, and the layout of the support structure. The FLAC3D numerical simulation software (a professional geotechnical engineering simulation tool) is subsequently used to simulate the stress‒strain state, deformation law, and stability of the rock surrounding roadways in each support scheme. Finally, on the basis of the numerical simulation results (such as the degree of roadway convergence, surrounding rock stress concentration coefficient, and support structure force), the optimal roadway support scheme is selected from the limited number of support design schemes through comparative analysis [17–19]. While, many other scholars try to obtain the optimal support from the perspective of stress and strain, such as, Li et al. [20] performed the field measurement and numerical simulation, the results reveal asymmetric stress and principal stress axis deflection in mining-damaged roadways. An integrated roof cutting and asymmetric support technology is proposed, whose effectiveness is verified via field monitoring.

Meanwhile, to investigate the influence of fault geometry and roadway cross-section shape on the collapse of roadways, some experiments were conducted, and the corresponding results were obtained. Wang et al. [21] conducted the uniaxial compression tests on double-hole specimens with pre-embedded joints were conducte, with DIC and AE used to monitor strain evolution and crack development. It can be found that circular twin roadways are more stable than square ones. Collapse stress drops to the minimum at α = 45°before rising as α increases from 0°to 90°. Numerical analysis supplements the tests, offering guidance for roadway construction in jointed rock masses. Mei et al. [22] developed a theoretical-numerical framework for transient P- and SV-wave responses of twin arbitrary-section tunnels via conformal mapping and complex variable theory, validated by limit cases. CASRock simulations further reveal static-dynamic coupled elastoplastic failure, offering references for deep multi-tunnel support design. In Yan’s studies [23], model tests and validated numerical simulations are adopted to explore structural mechanical responses, surrounding rock earth pressure and existing lining safety. Parametric analysis shows rising buffer stiffness increases lining internal force, and a 10 MPa elastic modulus buffer is recommended. The optimized scheme adapts to 1.46–3.06 m ultra-small clearance and cuts concrete consumption, providing references for mitigating construction disturbance to adjacent tunnels. These studies provide novel insights for research on roadway stability.

However, through the systematic analysis and induction of the research results of existing scholars, there are obvious limitations in the process used to determine the optimal roadway support scheme in most previous studies. Specifically, the process of obtaining the optimal support scheme is not scientific or rigorous enough. Owing to the lack of a unified and objective evaluation standard and quantitative analysis method, the process of selecting the support scheme and the final optimal support design determined are influenced strongly by the subjective experience and professional judgement of researchers. This subjectivity often leads to deviations in the optimal support scheme, making it unable to fully adapt to the actual geological conditions of the roadway and thus affecting the stability of the roadway support and the rationality of the investment.

To overcome the above limitations of existing studies and improve the scientific rigor, objectivity, and reliability of determining the optimal roadway support scheme, this study combines the firefly algorithm (FFA) with FLAC3D numerical simulation to establish a new method for determining the optimal roadway support design. In this method, the firefly algorithm, which has the advantages of strong global search ability, fast convergence, and easy implementation, is used as the core optimization method to dynamically update and optimize the key support parameters (such as the anchor rod length, anchor cable spacing, and grouting area). To verify the effectiveness and practicality of the proposed method, the optimized roadway support design scheme was applied to actual engineering practice. During the engineering implementation process, key monitoring indicators, such as the deformation of the rock surrounding roadways and water seepage volume, were recorded in detail. A comparison of the actual monitoring data with the numerical simulation results and the design requirements reveals that the deformation of the rock surrounding roadways with the optimized support scheme is within the allowable range, the support structure is in a stable stress state, and there is no obvious water seepage. These results indicate that the proposed method combining the FFA and FLAC3D numerical simulation is reliable and workable and can provide an effective technical reference for the optimal design of roadway supports under complex geological conditions.

2. Determining the optimal roadway support by using firefly algorithm

2.1. Firefly algorithm

The firefly algorithm (FFA) is a swarm intelligence optimization algorithm inspired by the flashing behaviour of fireflies that Xin-She Yang proposed in 2010 [24]. Its core logic is that fireflies move towards brighter counterparts, with brightness tied to the objective function value and attractiveness decaying with distance. The FFA has several advantages: first, it features a simple and intuitive mechanism with few adjustable parameters, making it easy to understand, implement and program. Second, it exhibits strong global search capability, especially for multimodal optimization problems, as its attraction mechanism allows fireflies to automatically cluster into subgroups and effectively avoid premature convergence to local optima. Third, it achieves a good balance between global exploration and local exploitation by adjusting randomness and absorption coefficients. The detailed process of FFA algorithm is described below.

Before the process is considered, it is critical to grasp the three foundational assumptions:

  • (1) All fireflies are unisex—any firefly will be attracted to all other brighter fireflies (no sex constraints).

  • (2) Attractiveness is proportional to brightness (brighter fireflies are more attractive) and decays exponentially with distance (attractiveness decreases as distance increases).

  • (3) The brightness of a firefly is determined by the objective function value: For maximization problems, the brightness I∝f(x) (higher f(x) = brighter). For minimization problems, the brightness I∝1/f(x) (lower f(x) = brighter).

The main procedure for the FFA is described as follows:

Step 1: Initialize the Algorithm Parameters

Define the core hyperparameters (adjustable based on the optimization problem). n: Number of fireflies (swarm size, typically 20–50). d: Dimension of the optimization problem (number of variables to optimize). α: Randomization parameter (step size, usually 0.01–1.0). β0: Initial attractiveness at distance 0 (base attractiveness, usually 1.0). γ: Light absorption coefficient (controls attractiveness decay, usually 0.1–10). max_iter: Maximum number of iterations (stopping criterion). Search range: [xmin, xmax] for each dimension.

Step 2: Initialize the Firefly Population

Randomly generate the initial position of each firefly in the search space:

Xi=(xi1,xi2,•••,xid),i=1,2,•••,n (1)

where xij ~ is uniform([xmin, xmax) (uniform random distribution).

Step 3: Calculate the Brightness and Attractiveness

Brightness calculation: For minimization, Ii=1/(1+f(Xi)) (division is avoided by 0 if f(Xi) =0). For maximization, Ii=f(Xi). (Note: f(Xi) is the objective function value of firefly i.) Distance between fireflies i and j: The Euclidean distance (most common) can be expressed as follows:

rij=∥Xi−Xj∥=∑k=1d(xik−xjk)2 (2)

Attractiveness calculation (decays with distance):

β(rij)=β0e−γrij2 (3)

Exponential decay ensures that attractiveness decreases rapidly with distance; γ controls the decay rate.

Step 4: Update the Firefly Positions

For each pair of fireflies i and j,

If Ij>Ii (firefly j is brighter), firefly i moves towards j:

Xi=Xi+β(rij)•(Xj−Xi)+α•(rand+0.5) (4)

β(rij)•(Xj−Xi): Attractive movement towards the brighter firefly. α•(rand+0.5): Random perturbation (exploration of the search space, rand∈[0,1]).

If Ij<Ii, Firefly i moves randomly (no attraction).

The brightest firefly (global optimal thus far) only moves randomly (to avoid local optimal values):

Xbest=Xbest+α•(rand−0.5) (5)

Step 5: Evaluate the New Positions

Recalculate the brightness Ii for all fireflies with updated positions, and update the global best firefly (brightest one).

Step 6: Check the Termination Criterion

If the number of iterations reaches max_iter or if the objective function value converges (no significant change for multiple iterations), stop the algorithm.

Otherwise, return to Step 3 and repeat.

2.2. Theory of fluid‒mechanical coupling in FLAC3D

The fluid‒mechanical coupling analysis in FLAC3D is grounded in the quasistatic Biot’s consolidation theory, which elaborates the bidirectional interactive mechanism between the seepage of single-phase fluid in porous geomaterials and the deformation/failure of the solid skeleton [25,26]. The numerical solution adopts the explicit finite difference method to solve the governing equations of solid mechanics and fluid flow alternately and to calculate the dynamic coupling of pore pressure and the stress‒strain field.

FLAC3D adopts the staggered explicit solution scheme for FM coupling, which solves the solid mechanical equilibrium equation and the fluid seepage control equation alternately in each time step without solving the coupled equations simultaneously, thus reducing the computational complexity and improving the efficiency:

  • (1) Mechanical step: The displacement, strain and stress of the solid skeleton are calculated at the current pore pressure, and the volumetric strain εv is updated;

  • (2) Hydraulic step: Based on the updated volumetric strain εv and porosity φ, the fluid mass conservation equation is solved to update the pore pressure field p;

  • (3) Iteration: The above two steps are repeated until the preset calculation time or convergence condition is reached (e.g., the variation in pore pressure and displacement is less than the threshold).

The time step of the coupling calculation must satisfy the stability condition for both the mechanical and hydraulic fields: the mechanical time step Δtm is controlled by the wave velocity of the solid medium, and the hydraulic time step Δtf is controlled by the seepage velocity and grid size. FLAC3D automatically synchronizes or adjusts the two time steps to avoid numerical oscillation and ensure calculation stability.

Key Assumptions of FLAC3D FM Coupling Theory

  • (1) The fluid is single phase, and seepage conforms to Darcy’s law (laminar flow and low seepage velocity);

  • (2) The deformation of the solid skeleton is small (small strain assumption), and the large deformation can be considered by enabling large-strain mode;

  • (3) The porous medium is fully saturated (unsaturated seepage must be realized by the unsaturated fluid module with the van Genuchten model);

  • (4) The coupling is quasistatic, and the inertial effect of fluid flow is ignored (dynamic seepage can be considered in special settings);

  • (5) The solid particles and fluid are slightly compressible (an incompressible fluid can be set by assigning a large Kf).

2.3. Optimizing the roadway support

Before the numerical simulation model is constructed, the centre point of the roadway should be determined, which is O in Figs 1–3. Then, the area around centre point O is divided into 10 parts evenly by 10 lines, and the lengths of the ten lines are l1,l2,•••,l10; on the basis of the coordinate (x0,y0) of the centre point O, the grouting area end point p1,p2,•••,p10 can be obtained as follows:

Fig 1. End point of the grouting area.

Fig 1

Fig 3. Grouting area in the numerical simulation model (FLAC3D).

Fig 3

{@l@xi=x0+licos((i−1)•36)yi=y0+lisin((i−1)•36)i=1,2,•••,10 (6)

Once ten end points p1,p2,•••,p10 of the grouting area are determined, a closed line can be obtained, as shown by the red dotted line in Fig 1. The closed line is a smooth curve. The smooth curve is obtained on the basis of ten end points p1,p2,•••,p10 and the cubic spline interpolation method.

The area within the range of the closed line and the roadway edge is subsequently obtained, as shown in Fig 1. For the numerical simulation model, when the zone is within the range of the area, the zone is grouped and considered the grouting area, as shown in Fig 2 and Fig 3.

Fig 2. Grouting area determined on the basis of the roadway edge and the closed line.

Fig 2

Moreover, to reduce the deformation of the roadway, cables are used, and the cable distribution is shown in Figs 4–6.

Fig 4. −arctan(Ha)≤β<0∘.

Fig 4

Fig 6. (180∘≤β<180∘+arctan(Ha).

Fig 6

As shown in Figs 4–6, the starting point of a cable is located at the edge of the roadway, and the length of a cable is L; then, the end point of a cable can be calculated on the basis of the cable length L, the starting point (x1,y1) of a cable and the inclination angle γ of a cable:

{@l@x2=x1+Lcosγy2=y1+Lsinγ (7)

Moreover, when the starting point (x1,y1) is fixed, the range of the inclination angle of the cable can be determined as follows:

When the starting point angle β is within the range of −arctan(Ha)≤β<0∘ (Fig 4), the inclination angle of the cable is within the range of −90∘≤γ≤90∘.

When the starting point angle β is within the range of 0∘≤β<180∘ (Fig 5), the range of the cable inclination angle is within the range of −arctan(b2x1a2y1)≤γ≤−arctan(b2x1a2y1)+180∘.

Fig 5. 0∘≤β<180∘.

Fig 5

When the starting point angle β is within the range of 180∘≤β<180∘+arctan(Ha) (Fig 6), the inclination angle of the cable is within the range of 90∘≤γ≤270∘.

The results of the analysis above reveal that the roadway water prevention and support variables have a length of ten l1,l2,•••,l10, a cable number, a cable length L (to reduce the iteration time and for the convenience of engineering practice, all the cable lengths are equal), a starting point for the cable (all the starting points should be on the edge of the roadway), and an inclination angle for the cable. The objective function can be expressed as follows:

f=Agrouting_area×Lall_cable_length (8)

Moreover, the roof and floor displacement should be less than or equal to 0.3 m, and the water flow into the roadway should be less than 0.2 m3/h. During the numerical simulation, if the roof and floor displacements are greater than 0.3 m or if the water flow into the roadway is greater than 0.2 m3/h, then the objective function value is set to infinity.

Therefore, the constraint condition and the objective function are determined, and the procedure for determining the optimal grouting area and cable distribution is as follows:

Step 1: Initialize the parameters of the firefly algorithm (FFA), in this paper, the population size is 50, randomization parameter α is 0.02, initial attractiveness at distance 0 β0 is 1.0, the light absorption coefficient γ is 2.0, the maximum number of iterations max_iter is 1000, and these hyperparameters were selected based on the reference [24]. And the roadway support parameters (Table 1).

Table 1. The initialized variables of supported parameters.

Variables Ranges
Grouting area (lengths of the ten lines are l1,l2,•••,l10) [2.7m, 10m]
Cable number [3,20]
Cable start points (x1,y1) The cable start points (x1,y1) should be on the edge of the roadway, it should be noted that the cable start points number would be changed when the cable number varied.
Cable angle γ starting point angle β is within the range of 0∘≤β<180∘, cable inclination angle is within the range of −arctan(b2x1a2y1)≤γ≤−arctan(b2x1a2y1)+180∘; starting point angle β is within the range of 180∘≤β<180∘+arctan(Ha), cable inclination angle is within the range of 90∘≤γ≤270∘. The number cable angle would be changed when the cable number varied.

Step 2: Generate initial solutions for numerical simulation on the basis of the predefined variable ranges and hyperparameters of the FFA.

Step 3: Generate the FLAC3D command stream using the grouting area parameters and cable parameters (number of cables, cable length, and cable position), and then execute the numerical simulation. On the basis of the simulation results (roof-floor relative displacement and roadway water inflow), and calculate the objective value via Equation 8 if the following constraints are met: roof−floor displacement ≤ 0.3 m, two-sided wall displacement ≤0.3 m, and roadway water inflow ≤ 0.2 m3/h. Otherwise, the objective value shall be assigned the value calculated by Equation 8 plus 1e7.

Step 4: Adjust the roadway support variables using the FFA update mechanism.

Step 5: Check the termination condition: if it is satisfied, terminate the numerical simulation and proceed to Step 6; if not, return to Step 2 for iterative optimization.

Step 6: Terminate the algorithm, and output the optimal results.

To implement the above process, Python scripting was adopted to fully automate and control the numerical simulation workflow using the FFA framework. The simulation iterates until the termination condition is met, after which the optimal grouting area and cable design parameters are determined. A numerical simulation case is presented to verify the effectiveness of the proposed method.

2.4. Example for verifying the proposed algorithm

To validate the proposed method, a numerical simulation was conducted, and the roadway of the Banji Coal Mine in Anhui, China, was selected as an example. The buried depth of roadway is 636 m, the height of the roadway was 3.2 m, the width of the roadway was 5.4 m, the speed of water flow into the roadway was 97 m3/h, the relative displacement between the roof and floor was 1.36 m, and the relative displacement of the two sides was 0.83 m. The underground water locations are shown in Fig 7. Water flows into roadways, and the large displacement of the roadways seriously threatens daily production.

Fig 7. Underground water location and geometry of the numerical simulation model.

Fig 7

To determine the optimal cable location and grouting area, a FLAC3D numerical simulation model was constructed, the dimension of the numerical is 43.2m × 1m × 34.4m, and 35550 grid points and 17500 zones included in the numerical simulation model. And the centre coordinate of the roadway was the location of point (0, 0); the mechanical parameters of the rock mass and grouting area are listed in Table 2, these mechanical parameters were obtained from laboratory tests. In the numerical simulation, the Mohr-Colomb model was used.

Table 2. Mechanical parameters of the rock mass and grouting area.

Density(kg/m3) Bulk modulus(GPa) Shear modulus (GPa) Cohesion(MPa) Friction angle (°) Tension(MPa)
Rock mass 2573 9.5 3.1 1.2 9.2 0.6
Grouting area 2690 10.7 3.7 5.3 31.3 2.3

The permeability coefficient of the rock mass is 7.9e-11 m/s, whereas the permeability coefficient of the grouting area is 4.5e-13 m/s, and the porosity of rock mass is 0.45 and the grouting area is 0.43, and the Biot coefficient for rock mass is 0.85, and the grouting area is 0.75, and the fluid bulk is 1.3GPa, these mechanical parameters were obtained from laboratory tests, and the roadway surrounding rock surface is set as zero pore water pressure drainage boundary. A constant water pressure of 2.3 MPa is applied to the bottom aquifer within 8 m height to simulate deep confined water recharge. The top, left and right boundaries of the numerical model are all defined as impermeable no-flow boundaries. Groundwater flows steadily from the high-pressure bottom water-rich zone to the roadway drainage surface, and the stable water inflow is finally calculated.

Furthermore, the mechanical parameters of the cables are listed in Table 3, these mechanical parameters were obtained from laboratory tests.

Table 3. Mechanical parameters of the cables.

Young’s modulus (Pa) 201e9 Yield stress (KN) 562 Grout friction angle (°) 35
Poisson’s ratio 0.2 Grout stiffness (Pa) 1.32e7 Pretension(KN) 100
Cross-sectional area (m2) 396e-6 Grout cohesion (Pa) 1.95e5 Exposed perimeter(m) 0.0705

To accurately determine the most reasonable reinforcement scheme for the target structure, the firefly algorithm (FFA) was adopted for systematic optimization calculations. Through multiple iterations and simulation analysis based on the FFA algorithm, the optimal grouting area and the corresponding reasonable cable distribution layout were finally obtained. These key optimization results, including the specific range for parameters such as the grouting area and the detailed geometric parameters of the cables (such as diameter, spacing, and arrangement angle), are clearly listed in Table 4, Table 5, Figs 8–12 for reference and verification. Specifically, the optimization results indicate that the total number of cables required for the reinforcement scheme is 11, and each cable has a uniform length of 7.28 m, which is determined on the basis of the structural depth and demand for reinforcement to ensure sufficient bonding force and reinforcement effect. For more comprehensive and detailed parameter information, such as the specific coordinates of the grouting area, the cross-sectional parameters of the cables, and the relevant technical specifications, readers can refer to Table 4 and Table 5, where all the detailed parameters are systematically sorted and presented.

Table 4. Grouting length.

Series number Value/m Series number Value/m
l1 5.32 l6 7.32
l2 4.90 l7 7.98
l3 5.54 l8 6.68
l4 5.67 l9 5.75
l5 6.16 l10 5.54

Table 5. Cable geometry parameters.

Series number Cable length/m Start point/m Incline angle/°
1 7.28 (2.7, −0.99) −10
2 7.28 (−2.7, −0.99) 191
3 7.28 (2.7, −0.33) −5
4 7.28 (−2.7, −0.34) 185
5 7.28 (2.65, 0.39) 13
6 7.28 (−2.65, 0.40) 166
7 7.28 (1.84, 1.68) 32
8 7.28 (−1.90, 1.62) 147
9 7.28 (1.22, 2.05) 63
10 7.28 (−1.22, 2.06) 117
11 7.28 (0.01, 2.29) 91

Fig 8. Outline of the grouting area.

Fig 8

Fig 12. Deformation of the roadway with a support.

Fig 12

Fig 9. Grouting area in the numerical simulation model.

Fig 9

Fig 10. Cable distribution.

Fig 10

The numerical simulation results are systematically presented in Fig 11 and Fig 12. According to the simulation data, four points were determined to record the displacement of roadway(Fig 13), the relative vertical displacement between the roof and floor of the roadway is 0.27 m, whereas the relative horizontal displacement of the two sidewalls reaches 0.23 m. Both values are less than the allowable displacement specified in the relevant engineering standards, demonstrating satisfactory deformation control. In addition, the simulated rate of water inflow into the roadway is 0.169 m3/h, which is also below the critical threshold of 0.2 m3/h. The minimum objection function value is 8315.50m2▪m. Meanwhile, the total computation time is 26.4h, the computation time is acceptable. Furthermore, to verify the steadiness of the proposed method, the process was repeated for 10 times, and the convergence history was illustrated in Fig 14.

Fig 11. Pore pressure distribution after grouting.

Fig 11

Fig 13. Deformation monitoring point.

Fig 13

Fig 14. The convergence history of 10 repeated numerical simulation.

Fig 14

As illustrated in Fig 14, the final objective function value are quite close, the mean objective function value in Fig 14 is 8316.53 and the standard deviation value is 0.71, which means that the 10 objective value is quite close, and the minimum value of 10 results is 8315.50.

To further verify the accuracy and reliability of the numerical simulation outcomes, the grouting reinforcement scheme and cable arrangement design adopted in the simulation have been implemented in field engineering practice. The field monitoring data reveal that the actual water inflow into the roadway is 0.13 m3/h, the measured relative displacement between the roof and floor is 0.25 m, and the relative displacement of the two sides is 0.20 m, the displacement and the water flow inflow roadway were recorded for 3 months, hence, the recorded data was reliable. These field test results are highly consistent with the numerical simulation predictions. Both the numerical simulation analysis and onsite engineering verification demonstrate that the support and treatment method proposed in this study is rational and reliable. Therefore, this method can be effectively popularized and applied in similar underground engineering practices.

3. Discussion

Groundwater poses a significant threat to engineering construction, particularly in underground roadway mining. It not only disrupts normal construction progress but also leads to intractable safety hazards and economic losses. On the one hand, high-velocity groundwater inflow into roadways can directly trigger water inrush disasters, submerging equipment, blocking construction passages, endangering personnel safety, and even causing heavy casualties in severe cases. On the other hand, long-term groundwater seepage into rock pores and fractures severely reduces the strength, stiffness and integrity of the surrounding rock, weakens its bearing capacity, and further induces large deformation, floor heave, roof collapse or rib spalling. These issues delay construction and increase project maintenance costs. Therefore, scientific and effective measures are urgently required to control groundwater hazards in underground engineering.

Given the severe harm caused by groundwater in roadway engineering, numerous scholars worldwide have conducted extensive experimental research, theoretical analysis and engineering practice on groundwater control and roadway support schemes. Valuable achievements have been accumulated through long-term investigations, and various technical measures—including grouting reinforcement, drainage, waterproof curtains and diverse support structures—have been developed and applied to resist groundwater intrusion and control roadway deformation. Nevertheless, obvious shortcomings remain in current applications. For instance, over-support can ensure roadway stability and prevent deformation or collapse due to insufficient strength, yet it consumes excessive materials, labor and funds, resulting in poor economic efficiency. Conversely, inadequate support strength that fails to match the actual roadway stress and surrounding rock mechanical properties easily leads to excessive deformation and even collapse, bringing serious safety risks and economic losses. In most existing literature, researchers typically propose several alternative support schemes and select the so-called optimal one based on experience or simple comparison. Such a selection process lacks systematic theoretical guidance and scientific quantitative analysis, leading to non-scientific and subjective results that can hardly meet the demands of complex underground roadway engineering.

Accordingly, to resolve the problems of unreasonable support design and inefficient water control in underground roadway engineering, and to achieve optimal support and water prevention parameters for safe and economical construction, this study combines the firefly algorithm (FFA) with FLAC3D numerical simulation to establish a comprehensive optimization model for roadway support and seepage control. First, key factors influencing roadway stability and groundwater seepage were identified via theoretical analysis, and optimization objectives (including minimizing support cost, maximizing roadway stability and reducing seepage) as well as constraint conditions (such as support strength limits, waterproofing requirements and construction feasibility) were defined. Then, the FFA was adopted to optimize the initial support and water control parameters, while FLAC3D was used to simulate stress distribution, deformation characteristics and seepage behavior under different parameter combinations to verify their rationality. Through iterative algorithm optimization and numerical simulation, the optimal parameter combination for roadway support and seepage prevention was ultimately determined.

To further validate the effectiveness and practicality of the proposed method, an engineering case was adopted for verification. The optimized parameters and waterproof measures, after numerical verification, were applied in field practice. Engineering monitoring data show that numerical results are highly consistent with actual site conditions, demonstrating that the method is reliable, feasible and of good engineering application value. This research provides scientific theoretical guidance and practical references for the design of support and seepage prevention systems in underground roadways.

However, this study has several limitations to be addressed and supplemented in future research. Only cable support and grouting reinforcement were considered here, whereas actual underground engineering involves various supports (e.g., shotcrete, steel arch, and combined support) with distinct features and applicable scenarios. Future work will explore more support types, optimize the combination of multiple supports and anti-seepage measures, and account for complex geological conditions (variable rock properties, high ground stress, complex groundwater distribution). This will enhance the comprehensiveness, rigor, and applicability of the optimization method, supporting safer, more efficient, and economical underground engineering construction.

4. Conclusions

To accurately determine the optimal grouting area and reasonable cable distribution scheme in geotechnical engineering support design, in this study, an optimal algorithm was integrated with numerical simulation technology, and a systematic technical method for determining optimal support parameters was established. The main conclusions of this paper are summarized as follows:

  • (1) The firefly algorithm (FFA) and FLAC3D numerical simulation software were combined to form a complete technical framework for determining the optimal support parameters of the grouting and cables. By comparing and analysing the simulation results, the optimal combination of support parameters for ensuring the stability of the rock mass and minimizing the engineering cost was ultimately identified.

  • (2) To verify the validity, reliability, and engineering applicability of the proposed method (combining the FFA algorithm and FLAC3D numerical simulation), a specific engineering example was selected for verification. The proposed method was applied to an example project for calculation and optimization, and the optimal grouting area parameters and cable distribution scheme suitable for the example project were obtained. The obtained optimal support parameters were applied to an actual engineering example, and the results of comparison reveal that the onsite monitoring data are in good agreement with the numerical simulation results, with small deviations and high consistency. These findings indicate that the method proposed for determining optimal support parameters is scientific, reliable, and feasible and can provide effective technical support and reference for the design and construction of similar geotechnical engineering support projects.

Supporting information

S1 Data. Minimal dataset 1.

(DOC)

pone.0357258.s001.doc (55.5KB, doc)
S2 Data. Minimal dataset 2.

(TXT)

pone.0357258.s002.txt (134.8KB, txt)

Data Availability

All relevant data are within the paper.

Funding Statement

The author(s) received no specific funding for this work.

References

  • 1.Yin H, Zhou W, LaMoreaux JW. Water inrush conceptual site models for coal mines of China. Environ Earth Sci. 2018;77(22). doi: 10.1007/s12665-018-7920-6 [DOI] [Google Scholar]
  • 2.Sui WH. Catastrophic mechanism and its prevention and control of seepage deformation and failure of mining rock mass Ⅱ: A review of water inrush from seam floor. J Eng Geol. 2022;30(6):1849–66. [Google Scholar]
  • 3.Sun W, Li W, Ning D. Current states, prediction and prevention suggestions for water hazard accidents in China’s coal mines. Coal Geol Exploration. 2023;51(12):185–94. [Google Scholar]
  • 4.Li B, Wang X, Liu Z, Li T. Study on multi-field catastrophe evolution laws of water inrush from concealed karst cave in roadway excavation: A case of Jiyuan coal mine. Geomatics, Natural Hazards and Risk. 2021;12(1):222–43. doi: 10.1080/19475705.2020.1870170 [DOI] [Google Scholar]
  • 5.Li B, Wu Q. Catastrophic evolution of water inrush from a water-rich fault in front of roadway development: A case study of the Hongcai coal mine. Mine Water Environ. 2019;38(2):421–30. doi: 10.1007/s10230-018-00584-z [DOI] [Google Scholar]
  • 6.Kang H, Yang J, Jiang P, Gao F, Li W, Li J, et al. Theory, technology and application of grouted bolting in soft rock roadways of deep coal mines. Int J Miner Metall Mater. 2024;31(7):1463–79. doi: 10.1007/s12613-024-2906-8 [DOI] [Google Scholar]
  • 7.Liu Q, Xue Y, Ma D, Li Q. Correction: Liu et al. failure characteristics of the water-resisting coal pillar under stress-seepage coupling and determination of reasonable coal pillar width. water 2023, 15, 1002. Water. 2023;15(10):1804. doi: 10.3390/w15101804 [DOI] [Google Scholar]
  • 8.Zhang Y, Tang S, Shi K. Risk assessment of coal mine water inrush based on PCA-DBN. Sci Rep. 2022;12(1):1370. doi: 10.1038/s41598-022-05473-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Qu Y, Xu M, Hou Y, Li C, Chen Y, Chen H, et al. Optimization of support parameters for large-section open-cut roadways in fully mechanized mining with large mining height. Appl Sci. 2025;15(8):4125. doi: 10.3390/app15084125 [DOI] [Google Scholar]
  • 10.Chen G, Guo M. Optimization of support parameters of auxiliary transport roadway in Lucun No.1 Mine. J Heilongjiang Univ Sci Technol. 2025;35(3):360–7. [Google Scholar]
  • 11.Mahdevari S, Hayati M. Finite-difference based response surface methodology to optimize tailgate support systems in longwall coal mining. Sci Rep. 2021;11(1):2321. doi: 10.1038/s41598-021-82104-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Xu Y, Chang Q, Yan X, Han W, Chang B, Bai J. Analysis and lessons of a mine water inrush accident resulted from the closed mines. Arab J Geosci. 2020;13(14). doi: 10.1007/s12517-020-05613-2 [DOI] [Google Scholar]
  • 13.Ye Z, Shoufeng T, Ke S, Xiamin T. An evaluation of the mine water inrush based on the deep learning of ISMOTE. Nat Hazards. 2023;117(2):1475–91. doi: 10.1007/s11069-023-05912-3 [DOI] [Google Scholar]
  • 14.Zheng Z, Xu H, Wang W, Zhang Q, Wang Y, Sun Q, et al. Seepage-stress combined experiment and damage model of rock in different loading and unloading paths. International Journal of Damage Mechanics. 2023;33(1):3–38. doi: 10.1177/10567895231193056 [DOI] [Google Scholar]
  • 15.Liu C, Chen C. A new apparatus for investigating stress, deformation and seepage coupling properties of rock fractures. J Cent South Univ Technol. 2011;18(5):1662–6. doi: 10.1007/s11771-011-0886-0 [DOI] [Google Scholar]
  • 16.Qiu C, Yang Y, Chen Y, Huang B, Shang R, Cai C, et al. Pore-fracture evolution and seepage behavior in sandstone under triaxial stress: Insights from real-time CT imaging. International J Rock Mechanics and Mining Sci. 2025;195:106292. doi: 10.1016/j.ijrmms.2025.106292 [DOI] [Google Scholar]
  • 17.Wu Q, Liu H, Dai B, Cheng L, Li D, Qin P. Influence of Base-angle bolt support parameters and different sections on overall stability of a roadway under a deeply buried high stress environment based on numerical simulation. Sustainability. 2023;15(3):2496. doi: 10.3390/su15032496 [DOI] [Google Scholar]
  • 18.Pu L, Liu Y, Cai Y, Sun Z, Zhou X. Study on active support parameters for surrounding rock with ultra-large span open-off cut in thick coal seam. Appl Sci. 2023;13(23):12804. doi: 10.3390/app132312804 [DOI] [Google Scholar]
  • 19.Li C, Nan H, Zhang W. Research and application of foam filling material in soft rock roadways. Appl Sci. 2025;15(2):520. doi: 10.3390/app15020520 [DOI] [Google Scholar]
  • 20.Li YL, Dai XL, Yang RS. Asymmetric large deformation in deep mining-damaged roadways: mechanism and collaborative control technology by means of destressing-bolting. Geomechanics and Geophysics for Geo-Energy and Geo-Resources. 2026;12(1):98. [Google Scholar]
  • 21.Wang F, Xie H, Zhou C, Wang Z, Li C. Combined effects of fault geometry and roadway cross-section shape on the collapse behaviors of twin roadways: An experimental investigation. Tunnelling and Underground Space Technol. 2023;137:105106. doi: 10.1016/j.tust.2023.105106 [DOI] [Google Scholar]
  • 22.Mei W, Pan P, Wang Z, Zhu Y, Xie Y, Wang Z. Transient response of deep-buried twin tunnels with arbitrary cross-sections under P- and SV-waves. Comput Geotechnics. 2026;194:108016. doi: 10.1016/j.compgeo.2026.108016 [DOI] [Google Scholar]
  • 23.Yan F, Qiu W, Zheng Y, Jiang S, Hu H, Gao G, et al. Experimental investigation and numerical analysis of mechanical behaviour of railway station reconstruction over twin tunnels. Tunnelling and Underground Space Technol. 2021;112:103918. doi: 10.1016/j.tust.2021.103918 [DOI] [Google Scholar]
  • 24.Yang XS. Firefly algorithm, stochastic test functions and design optimisation. Int J Bio-Inspired Comput. 2010;2(2):78. doi: 10.1504/ijbic.2010.032124 [DOI] [Google Scholar]
  • 25.Itasca Consulting Group. FLAC3D Version 9.0 Theory Manual: Fluid-Mechanical Coupling. Itasca Consulting Group Inc. 2025. [Google Scholar]
  • 26.Giraldo Zapata VM, Botero Jaramillo E, Ossa Lopez A. Implementation of a model of elastoviscoplastic consolidation behavior in Flac 3D. Comput Geotechnics. 2018;98:132–43. doi: 10.1016/j.compgeo.2017.11.011 [DOI] [Google Scholar]

Decision Letter 0

Zhenhua Li

23 Jul 2026

Dear Dr. xu,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Sep 21 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

  • A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

As the corresponding author, your ORCID iD is verified in the submission system and will appear in the published article. PLOS supports the use of ORCID, and we encourage all coauthors to register for an ORCID iD and use it as well. Please encourage your coauthors to verify their ORCID iD within the submission system before final acceptance, as unverified ORCID iDs will not appear in the published article. Only  the individual author can complete the verification step; PLOS staff cannot  verify ORCID iDs on behalf of authors.

We look forward to receiving your revised manuscript.

Kind regards,

Zhenhua Li

Academic Editor

PLOS One

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please note that PLOS One has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, we expect all author-generated code to be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse.

3. Please update your submission to use the PLOS LaTeX template. The template and more information on our requirements for LaTeX submissions can be found at http://journals.plos.org/plosone/s/latex.

4. Thank you for stating the following in your Competing Interests section:

“no”

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

5. Please provide a complete Data Availability Statement in the submission form, ensuring you include all necessary access information or a reason for why you are unable to make your data freely accessible. If your research concerns only data provided within your submission, please write "All data are in the manuscript and/or supporting information files" as your Data Availability Statement.

6. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

7. Thank you for stating the following in the Competing Interests section:

“no”

We note that one or more of the authors are employed by a commercial company: “China Coal Underground Space Technology (Shenzhen) Co., Ltd.”

1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

8. Please upload a new copy of Figure 3 as the detail is not clear. Please follow the link for more information:  https://journals.plos.org/plosone/s/figures

9. Please ensure that you refer to Figure 4 in your text as, if accepted, production will need this reference to link the reader to the figure.

10. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

Additional Editor Comments:

You will see from the referees' comments that additional information needs to be provided, and we ask that this be provided, before we consider you manuscript further.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: Yes

**********

Reviewer #1: This study integrates firefly algorithm (FFA) with FLAC3D numerical simulation. Taking Banji Coal Mine roadway as the research object, FFA is adopted to optimize grouting range and anchor cable layout. Field application verifies that simulation results match practical data well. The study is meaningful and interesting, however, some questions and revision should be taken before the acceptance.

(1)In this study, only one Coal Mine was taken as an example, the proposed method can be suitable for other coal mine

(2)The FFA algorithm was used to find the optimal parameters for water prevention, why this algorithm was selected? please explain the reasons.

(3)In this paper, the FFA algorithm and FLAC3D numerical simulation combined by Python language, are you sure the Python can be used for FLAC3D numerical simulation?

(4)The obtained numerical simulation results were applied to engineering practice, the difference between numerical simulation and engineering practice, how do you determine the mechanical parameters of rock mass in numerical simulation.

(5)How do you determine the range of cable length in the manuscript.

This paper can be accepted after minor revision.

Reviewer #2: The manuscript titled “Optimal Parameters of Roadway Supports for Water Prevention based on the FFA Algorithm” combines the firefly algorithm (FFA) with FLAC3D fluid–mechanical simulation to determine the optimal grouting area and cable layout of a water-bearing roadway, and the optimized scheme was applied to the Banji Coal Mine. The topic is of practical value in mining engineering, and the field implementation is a strength that many optimization studies do not have. However, several material parameters and boundary conditions are physically inadmissible or contradictory, and most of the information needed to reproduce the work is missing, so the results cannot be assessed in the current version. Please see my comments below:

1. In Table 1, the bulk modulus (3.1 GPa) is much smaller than the shear modulus (9.5 GPa), which gives a negative Poisson’s ratio for the rock mass (ν = −0.26). The grouting area gives ν = −0.24. This is not possible for coal measure rock. If the two columns are exchanged, ν ≈ 0.35, which is reasonable. Please check the laboratory data, correct Table 1 and re-run all the simulations, because every mechanical result depends on these values.

2. The seepage boundary condition is contradictory. The text says the model boundary is a waterproof boundary, the water pressure at the roadway edge is 0 MPa, and the flow is steady, but Fig. 3 shows a water pressure of 2.3 MPa applied at the bottom over 8 m. If all outer boundaries are impermeable and the roadway is a drain, the steady-state inflow should be zero, and the reported 0.169 m³/h cannot be obtained. Please give the complete hydraulic boundary conditions and make the text consistent with Fig. 3.

3. The dimension of the optimization problem is not consistent. Section 2.3 lists ten grouting lengths, the cable number, the cable length, and the starting point and inclination angle of each cable, which is at least 34 variables for 11 cables, but Step 1 of the same section says the variable is ten-dimensional. Please give the variable vector, its dimension, and the search range of each variable in a table.

4. The parameters of the FFA are not reported. The population size, α, β₀, γ and max_iter actually used are not given, only the general ranges in Section 2.1, and there is no convergence curve, iteration number, total number of FLAC3D calls, or computation time. As FFA is a stochastic algorithm, please add the hyperparameters, the convergence history, and several independent runs with the mean and scatter of the best objective value.

5. The numerical model is not described. Please add the model dimensions, element size, number of zones, and a mesh sensitivity analysis. The porosity, Biot coefficient and fluid bulk modulus required by Equations (10)–(13) are not given, and the unit of the permeability coefficient (7.9e−11 and 4.5e−13) is missing, so these values cannot be interpreted.

6. In Section 2.3, Step 3, the objective value of an infeasible solution is set to infinity. If most of the initial population is infeasible, all fireflies have the same brightness, the attraction term disappears, and the algorithm degenerates into a random search. Please report the feasible proportion of the initial population, and consider a graded penalty function or a repair operator.

7. Section 2.2 repeats the FLAC3D theory manual (Reference [21]) at too much length. Equations (6)–(13) are presented but the parameters they introduce (α, ϕ, K_f, M_b) are never assigned values or used afterwards. Please shorten this section and keep only the formulation and parameters actually adopted, with the source clearly cited.

8. The grouting area is described by ten rays at 36° interval, but the reason for choosing ten is not explained and no sensitivity analysis of the number of control points is given. The optimized boundary in Table 3 (4.90–7.98 m) is close to an ellipse, so it is not clear how much the ten parameters contribute. Please also discuss whether such a shape can actually be constructed by borehole grouting.

9. The water inflow decreases from 97 m³/h to 0.13 m³/h, about 750 times, while the permeability ratio of the rock mass and the grouting area is only about 176. Please give a simple consistency check of this reduction against the grouting geometry.

10. In Section 2.4, “The roadway height was less than 636 m, the height of the roadway was 3.2 m” is contradictory in one sentence. The first value should be the burial depth.

11. In Table 2, “Yield tension (Pa) 1420e6” is a stress, not a tension. In FLAC3D the cable yield tension is a force; with the cross-sectional area of 396e−6 m² it should be about 562 kN. The pretension and the exposed perimeter of the cable are also not given.

12. In Section 2.1, the FFA is said to be proposed by Yang in 2008, but Reference [20] is dated 2010. Reference [1] and several other entries are also not in the correct format, and References [12], [13], [18] and [19], which are about Schmidt hammer tests and DEM parameter calibration, are not related to roadway support. Please check the whole reference list.

13. In Section 2.3, “As shown in Fig., the starting point of a cable …” the figure number is missing. In Section 2.4, Figs. 5–8 are listed but Fig. 4 is omitted, and the caption of Fig. 1(b) is separated from the figure.

14. Some references should be added to enrich the research background. For example, Combined effects of fault geometry and roadway cross-section shape on the collapse behaviors of twin roadways: an experimental investigation, Failure mechanism and support technology of weakly cemented soft rock roadway in water-rich environment, Asymmetric large deformation in deep mining-damaged roadways: mechanism and collaborative control technology by means of destressing-bolting.

15. The English of the manuscript should be polished by a professional editor. Several sentences are repeated word by word, and some expressions are not standard.

**********

what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review?  For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures

You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation.

NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications.

PLoS One. 2026 Oct 1;21(10):e0357258. doi: 10.1371/journal.pone.0357258.r002

Author response to Decision Letter 1


10 Aug 2026

response to editor

1. Please upload a Response to Reviewers letter which should include a point by point response to each of the points made by the Editor and / or Reviewers. (This should be uploaded as a 'Response to Reviewers' file type.) Please follow this link for more information: http://blogs.PLOS.org/everyone/2011/05/10/how-to-submit-your-revised-manuscript/

Response: the response to reviewers letter include a point by point response to each points made by editor and reviewers has be uploaded.

2. It appears that your ORCiD iD has not been validated in your Editorial Manager account and we are unable to proceed until that step is complete.

To validate your ORCiD iD in Editorial Manager, please follow the steps below:

a. In your Editorial Manager account, please go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCiD field.

b. This link will take you to the ORCiD site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

For additional instructions, please watch the following video for a step-by-step demonstration: https://www.youtube.com/watch?v=_xcclfuvtxQ

Response: the ORCID has been validated.

3. Please provide a complete Data Availability Statement in the submission form, ensuring you include all necessary access information or a reason for why you are unable to make your data freely accessible. If your research concerns only data provided within your submission, please write "All data are in the manuscript and/or supporting information files" as your Data Availability Statement.

Response: data statement “All data are in the manuscript and/or supporting information files” was added.

4. It is important that you include a cover letter with your manuscript. Please ensure that this letter is addressed specifically to PLoS ONE. Please also include

* why this manuscript is suitable for publication in PLoS ONE.

* how does your paper provide a worthwhile addition to the scientific literature?

* how does your paper relate to previously published work?

* which types of scientists do you believe will be most interested in your study?

Response: the cover letter was revised.

5. Please amend the title either on the online submission form or in the manuscript so that they are identical.

Response: the title on the online submission form was revised.

6. Thank you for stating the following in your Competing Interests section:

“no”

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

Response: the competing interests was revised.

7. Thank you for stating the following in the Competing Interests section:

“no”

We note that one or more of the authors are employed by a commercial company: “China Coal Underground Space Technology (Shenzhen) Co., Ltd.”

1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

Response: The commercial company was deleted.

Response to reviewers:

Reviewer #1: This study integrates firefly algorithm (FFA) with FLAC3D numerical simulation. Taking Banji Coal Mine roadway as the research object, FFA is adopted to optimize grouting range and anchor cable layout. Field application verifies that simulation results match practical data well. The study is meaningful and interesting, however, some questions and revision should be taken before the acceptance.

Response: Thanks for your comments, based on your comments, the page number and line number were added in the manuscript, meanwhile, the exact changes made in the revise manuscript, including corresponding page and line numbers were given.

(1)In this study, only one Coal Mine was taken as an example, the proposed method can be suitable for other coal mine

Response: Thanks for your comments, the proposed method can be applied to other coal mine, in the manuscript, once the shape of roadway and rock mechanic parameters were determined, the optimal grouting area and cable distribution can be determined based on the proposed method.

(2)The FFA algorithm was used to find the optimal parameters for water prevention, why this algorithm was selected? please explain the reasons.

Response: Thanks for your comments, many other optimization algorithm can be also used in the manuscript. In our manuscript, the FFA algorithm was selected, the FFA has several advantages: first, it features a simple and intuitive mechanism with few adjustable parameters, making it easy to understand, implement and program. Second, it exhibits strong global search capability, especially for multimodal optimization problems, as its attraction mechanism allows fireflies to automatically cluster into subgroups and effectively avoid premature convergence to local optima. Third, it achieves a good balance between global exploration and local exploitation by adjusting randomness and absorption coefficients. The detailed process of FFA algorithm is described below. The merits of the FFA algorithm was given in part 2.1 of the manuscript.

(3)In this paper, the FFA algorithm and FLAC3D numerical simulation combined by Python language, are you sure the Python can be used for FLAC3D numerical simulation?

Response: Thanks for your comments, the FLAC3D has two secondary development language, one is the FISH language and the other one is the Python script language, the Python language was supported since Version 5.0, which can be seen the manual of FLAC3D.

(4)The obtained numerical simulation results were applied to engineering practice, the difference between numerical simulation and engineering practice, how do you determine the mechanical parameters of rock mass in numerical simulation.

Response: Thanks for your comments, in our manuscript, the rock mass mechanical parameters were obtained from physical tests, and then the rock mass mechanical parameters were calibrated based on the rock mass grade. It should be noted that numerical simulation results agree well with engineering practice, which indicated that the rock mass mechanical parameters applied to numerical simulation model is reliable.

(5)How do you determine the range of cable length in the manuscript.

Response: Thanks for your comments, in our manuscript, the range of cable length is within the range of 0.1m and 10m, in fact, wider of the range cable length can be more capable to obtain the optimal cable length, to reduce the iteration time, hence, the range (0.1, 10) was selected.

This paper can be accepted after minor revision.

Response: Thank you again. There is no questions needed to be answered.

Reviewer #2: The manuscript titled “Optimal Parameters of Roadway Supports for Water Prevention based on the FFA Algorithm” combines the firefly algorithm (FFA) with FLAC3D fluid–mechanical simulation to determine the optimal grouting area and cable layout of a water-bearing roadway, and the optimized scheme was applied to the Banji Coal Mine. The topic is of practical value in mining engineering, and the field implementation is a strength that many optimization studies do not have. However, several material parameters and boundary conditions are physically inadmissible or contradictory, and most of the information needed to reproduce the work is missing, so the results cannot be assessed in the current version. Please see my comments below:

Response: Thanks for your comments, based on your comments, the page number and line number were added in the manuscript, meanwhile, the exact changes made in the revise manuscript, including corresponding page and line numbers were given.

1. In Table 1, the bulk modulus (3.1 GPa) is much smaller than the shear modulus (9.5 GPa), which gives a negative Poisson’s ratio for the rock mass (ν = −0.26). The grouting area gives ν = −0.24. This is not possible for coal measure rock. If the two columns are exchanged, ν ≈ 0.35, which is reasonable. Please check the laboratory data, correct Table 1 and re-run all the simulations, because every mechanical result depends on these values.

Response: Thanks for your comments, it is our mistakes, the two columns should be exchanged, our numerical simulation are correct, there exist some mistakes when writing the manuscript, there is no need to re-run all the simulations. The changes were highlighted by red color, which can be seen in the revised manuscript.

2. The seepage boundary condition is contradictory. The text says the model boundary is a waterproof boundary, the water pressure at the roadway edge is 0 MPa, and the flow is steady, but Fig. 3 shows a water pressure of 2.3 MPa applied at the bottom over 8 m. If all outer boundaries are impermeable and the roadway is a drain, the steady-state inflow should be zero, and the reported 0.169 m³/h cannot be obtained. Please give the complete hydraulic boundary conditions and make the text consistent with Fig. 3.

Response: Thanks for your comments, based on your comments, the model boundary of the numerical simulation was changed. In our manuscript, the roadway surrounding rock surface is set as zero pore water pressure drainage boundary. A constant water pressure of 2.3 MPa is applied to the bottom aquifer within 8 m height to simulate deep confined water recharge. The top, left and right boundaries of the numerical model are all defined as impermeable no-flow boundaries. Groundwater flows steadily from the high-pressure bottom water-rich zone to the roadway drainage surface, and the stable water inflow is finally calculated. The changes were highlighted by red color, which can be seen in the revised manuscript.

3. The dimension of the optimization problem is not consistent. Section 2.3 lists ten grouting lengths, the cable number, the cable length, and the starting point and inclination angle of each cable, which is at least 34 variables for 11 cables, but Step 1 of the same section says the variable is ten-dimensional. Please give the variable vector, its dimension, and the search range of each variable in a table.

Response: Thanks for your comments, based on your comments, the initialized parameters and ranges were given in Table 1, the changes were highlighted by red color, and the changed can be seen in the revised manuscript.

4. The parameters of the FFA are not reported. The population size, α, β₀, γ and max_iter actually used are not given, only the general ranges in Section 2.1, and there is no convergence curve, iteration number, total number of FLAC3D calls, or computation time. As FFA is a stochastic algorithm, please add the hyperparameters, the convergence history, and several independent runs with the mean and scatter of the best objective value.

Response: Thanks for your comments, based on your comments, in this paper, for the FFA algorithm, the parameters were set as follows: Initialize the parameters of the firefly algorithm (FFA), in this paper, the population size is 50, randomization parameter is 0.02, initial attractiveness at distance 0 is 1.0, the light absorption coefficient is 2.0, the maximum number of iterations is 1000, the changes were added in the revised manuscript. Meanwhile, based on your comments, the convergence history, and the mean and scatter of the best objective value was given in the manuscript.

5. The numerical model is not described. Please add the model dimensions, element size, number of zones, and a mesh sensitivity analysis. The porosity, Biot coefficient and fluid bulk modulus required by Equations (10)–(13) are not given, and the unit of the permeability coefficient (7.9e−11 and 4.5e−13) is missing, so these values cannot be interpreted.

Response: Thanks for your comments, based on your comments, the numerical simulation model is described, the model dimensions, element size, number of zones were given, meanwhile, the porosity, Biot coefficient and fluid bulk modulus are given, which can be seen in the revised manuscript.

6.In Section 2.3, Step 3, the objective value of an infeasible solution is set to infinity. If most of the initial population is infeasible, all fireflies have the same brightness, the attraction term disappears, and the algorithm degenerates into a random search. Please report the feasible proportion of the initial population, and consider a graded penalty function or a repair operator.

Response: Thanks for your comments, if the condition is not satisfied, the objective value shall be assigned the value calculated by Equation (16) plus 1e7.

7. Section 2.2 repeats the FLAC3D theory manual (Reference [21]) at too much length. Equations (6)–(13) are presented but the parameters they introduce (α, φ, K_f, M_b) are never assigned values or used afterwards. Please shorten this section and keep only the formulation and parameters actually adopted, with the source clearly cited.

Response: Thanks for your comments, based on your comments, this part was shorten, which can be seen in the revised manuscript.

8. The grouting area is described by ten rays at 36° interval, but the reason for choosing ten is not explained and no sensitivity analysis of the number of control points is given. The optimized boundary in Table 3 (4.90–7.98 m) is close to an ellipse, so it is not clear how much the ten parameters contribute. Please also discuss whether such a shape can actually be constructed by borehole grouting.

Response: Thanks for your comments, as a matter of fact, more accuracy of the numerical results would be obtained if more end points selected. If more end points were used, then more variables needed to be determined during the process, in our manuscript, ten end points were determined based on the experience, meanwhile, the end points were connected by cubic spline interpolation method, the closed line is smooth, and the results were acceptable. However, the actual grouting depth of roadways after grouting can be comprehensively determined by construction parameters, on-site verification and non-destructive testing. The slurry diffusion range is preliminarily judged according to the final drilling depth, changes of grouting pressure and flow rate, and slurry consumption. After the slurry solidifies, core sampling and borehole inspection are conducted through inspection holes arranged between grouting holes to directly detect the maximum effective filling depth of slurry veins. Meanwhile, geophysical methods such as ground penetrating radar and acoustic wave detection are adopted to define the grouting reinforcement boundary, and the actual effective grouting depth of roadways is finally confirmed by mutual verification of various data. I hope my answer can meet your satisfactory.

9. The water inflow decreases from 97 m³/h to 0.13 m³/h, about 750 times, while the permeability ratio of the rock mass and the grouting area is only about 176. Please give a simple consistency check of this reduction against the grouting geometry.

Response: Thanks for your comments, The discrepancy is mainly caused by the change of seepage flow channel geometry after grouting. Grouting not only reduces rock mass permeability, but also greatly narrows, blocks and disconnects original dominant seepage paths. The actual seepage cross-sectional area decreases sharply, leading to a far larger multiple of water inflow reduction than the permeability ratio, which is consistent with the field grouting geometric characteristics. Meanwhile, the numerical simulation was repeated, the numerical simulation results were right.

10. In Section 2.4, “The roadway height was less than 636 m, the height of the roadway was 3.2 m” is contradictory in one sentence. The first value should be the burial depth.

Response: Thanks for your comments, the contradictory part was revised, which can be seen in the revised manuscript.

11. In Table 2, “Yield tension (Pa) 1420e6” is a stress, not a tension. In FLAC3D the cable yield tension is a force; with the cross-sectional area of 396e−6 m² it should be about 562 kN. The pretension and the exposed perimeter of the cable are also not given.

Response: Thanks for your comments, the unit of yield tension is revised, the pretension and the exposed perimeter of the cable are added in the manuscript.

12. In Section 2.1, the FFA is said to be proposed by Yang in 2008, but Reference [20] is dated 2010. Reference [1] and several other entries are also not in the correct format, and References [12], [13], [18] and [19], which are about Schmidt hammer tests and DEM parameter calibration, are not related to roadway support. Please check the whole reference list.

Response: Thanks for your comments, FFA is said to be proposed by Yang in 2010, which can be seen in the revised manuscript, and references [12], [13], [18] and [19] were deleted, and whole reference list were checked.

13. In Section 2.3, “As shown in Fig., the starting point of a cable …” the figure number is missing. In Section 2.4, Figs. 5–8 are listed but Fig. 4 is omitted, and the caption of Fig. 1(b) is separated from the figure.

Response: Thanks for your comments, in section 2.3, the figure number is added. Meanwhile, the Fig.4 is added. The Fig. 1 in the manuscript was revised.

14. Some references should be added to enrich the research background. For example, Combined effects of fault geometry and roadway cross-section shape on the collapse behaviors of twin roadways: an experimental investigation, Failure mechanism and support technology of weakly cemented soft rock roadway in water-rich environment, Asymmetric large deformation in deep mining-damaged roadways: mechanism and collaborative control technology by means of destressing-bolting.

Response: Thanks for your comments, based on your comments, some references were added to enrich the research background.

15. The English of the manuscript should be polished by a professional editor. Several sentences are repeated word by word, and some expressions are not standard.

Response: Thanks for your comments, based on your comments, our manuscript was polished by American Journal Expert, the editing proof was attached. The changed part was highlighted by red color, which can be seen in the revised manuscript.

Attachment

Submitted filename: Response to reviewers.doc

pone.0357258.s004.doc (54KB, doc)

Decision Letter 1

Zhenhua Li

16 Aug 2026

Optimal Parameters of Roadway Supports for Water Prevention based on the FFA Algorithm

PONE-D-26-35180R1

Dear Dr. xu,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Zhenhua Li

Academic Editor

PLOS One

Additional Editor Comments (optional):

accept

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions??>

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: Yes

**********

Reviewer #1: All reviewer comments have been satisfactorily resolved. The manuscript is now ready for acceptance.

Reviewer #2: I have no further comments on this manuscript, and this manuscript can be accepted for publication after check the grammar issues.

**********

what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review?  For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: Yes:  yes

Reviewer #2: No

**********

Acceptance letter

Zhenhua Li

PONE-D-26-35180R1

PLOS One

Dear Dr. xu,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS One and supporting open access.

Kind regards,

PLOS One Editorial Office Staff

on behalf of

Professor Zhenhua Li

Academic Editor

PLOS One

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Data. Minimal dataset 1.

    (DOC)

    pone.0357258.s001.doc (55.5KB, doc)
    S2 Data. Minimal dataset 2.

    (TXT)

    pone.0357258.s002.txt (134.8KB, txt)
    Attachment

    Submitted filename: Response to reviewers.doc

    pone.0357258.s004.doc (54KB, doc)

    Data Availability Statement

    All relevant data are within the paper.


    Articles from PLOS One are provided here courtesy of PLOS

    RESOURCES