Landslide Susceptibility Assessment Method during the Construction of Highways Based on the Index Complexity Algorithm

Abstract

Landslides represent the most destructive and prevalent geological hazards along mountainous highways, severely imperiling the construction and maintenance of road infrastructure. To mitigate risks associated with high slopes during construction, a systematic evaluation of landslide susceptibility is imperative. This study introduces an assessment method developed over three years of engineering practice, integrating ten parameters that are intricately linked to construction scale, geological conditions, and engineering design. The method innovatively employs the Index Complexity Algorithm (ICA) to ascertain the weight distribution of the parameters, thereby diminishing the impact of subjective biases in qualitative assessments and enhancing the objectivity and precision of the evaluation. Utilizing the slope in China as a case study, the paper meticulously demonstrates the application of the assessment method. A comprehensive evaluation of the slope’s geological context, construction scale, and design rationality by the ICA algorithm yields a quantified risk score for the slope’s potential hazards. The findings indicate that the slope is classified as high risk (Grade III) during highway construction, necessitating the implementation of risk mitigation measures such as prestressed anchor cables and grouting anchorage. Beyond offering a novel methodological approach to landslide risk assessment, the method significantly contributes to the sustainable construction and operation of mountainous highways. Anticipated refinements in the assessment process and the parameter are poised to augment the method’s efficacy in slope engineering safety management, thereby bolstering the long-term stability and environmental sustainability of mountain highways.

1. Introduction

Landslides represent a serious threat to human lives and can cause serious socioeconomic losses and environmental degradation [1,2]. These natural hazards also have the potential to severely damage critical highway infrastructure (CHI) [3,4], resulting in catastrophic disasters for people, the environment, and the economy [5,6].

Transportation corridors through mountainous areas often need to cut slopes to accommodate highway and railway alignments [7,8]. Slopes in some places become unstable and then cause hazards such as rock falls, slides, and debris and so on, potentially breaking or blocking highways and railway tracks (Figure 1). It is thus important for highway construction to assess the stability of slopes along both sides of the highway, which is directly related to the design and implementation of disaster prevention and mitigation measures [9,10,11]. In recent years, with the development of technology, the research methods in this field have shown a trend of diversification [12,13,14,15].

In the evaluation of landslide risk, traditional analytical methods such as the Analytic Hierarchy Process (AHP) and the Fuzzy Comprehensive Evaluation method rely heavily on expert experience and knowledge, which makes them somewhat subjective and limits their applicability [16,17]. In recent years, data-driven models such as Logistic Regression [18], Support Vector Machines (SVMs) [19], and Random Forests have been increasingly used for landslide risk assessment [20], but the results of these evaluations are largely dependent on sample data and model parameters. In addition to the characteristics and application research of the above evaluation methods, another main aspect of geological disaster assessment research is the selection of the evaluation unit [21]. The definition or selection of the evaluation unit for susceptibility and hazard evaluation should consider the research scale and the model used [11]. Many studies on landslide assessment in various research areas at home and abroad have been carried out using multivariate statistical methods and machine learning methods based on multivariate statistics, and these studies mostly use grid units for large-area evaluation research [22]. In practical engineering applications, accurately completing the risk assessment of individual landslides is also very important [23].

Due to the complex environmental conditions in mountainous areas [24], highway construction in China is often threatened by different geological hazards, especially landslides. Many factors are related to landslide hazards, such as geologic conditions, scale of slope, and construction environment. From 2015 to 2018, the authors were funded by the Ministry of Transport of Peoples Republic of China (MOTPRC) to develop a method for assessing the hazards induced by slope damage during highway construction (MOTPRC, 2015) [25]. After that, an assessment method was developed for the stability evaluation of slopes along highways. In this assessment method, an Index Complexity Algorithm (ICA) was proposed by using a weight coefficient distribution in the evaluation index system with 10 parameters. These 10 parameters are related to the construction scale, geological factors, and design and construction scheme. The method was developed to reduce the influence of subjectivity during qualitative weight distributions, which attracted widespread attention because of successful applications in many provinces of China, such as Guizhou province, Guangdong province, Yunnan province, and Fujian province. The assessment method helps prevent landslide-induced displacement and transportation disruptions, ensuring the safety and stability of communities.

Furthermore, landslides pose not only a threat to human life but can also cause long-term environmental damage, affecting ecosystem balance and biodiversity. By implementing the landslide susceptibility assessment method proposed in this study, potential landslide risks can be identified and mitigated during the design and construction phases, thereby promoting the resilience and sustainability of the environment.

2. Material and Methods

2.1. Study Area

The study area is located in the eastern slope of Guizhou plateau and has the following features: landform type is structural erosion-denudation mountain landform, the area relief is larger, surface erosion, erosion, near 663~720 m, relative height difference of 57 m, subgrade axis through elevation between approximately 692 and 713 m, relative height difference of 21 m, road surface for dry land and barren hills, vegetation development.

The original support design of the landslide is a grade 4 slope with a slope ratio of 1:0.75~1:1.25. The slope is supported by slurry stone retaining wall, and some slope is protected by frame grass planting. The original appearance of the slope and the site situation of the slope after the first reinforcement are shown in Figure 2.

The landslide area is covered with silty clay, and the underlying bedrock is the fully weathered layer of the Middle to Upper Silurian Wengxiang Group sandstone interbedded with mudstone, with local areas of strong weathering. During the initial excavation phase of the construction, small-scale shallow collapses occurred multiple times.

2.2. Hazard Features of Highway-Slope

It is well known that slope hazards during highway construction often result from a lack of geological data for slope stability evaluation and the limitation of the model for simulating the real geological world [26].

In the assessment method presented in this paper, a four-step process for hazard analysis was proposed (Figure 3), including hazard identification, hazard analysis, hazard assessment, and hazard management.

In the stage of hazard source identification, the hazard risk needs to be analyzed. It can be expressed quantitatively as Risk = Hazard × Potential values of Loss.

During hazard analysis, two analytical approaches are involved, including qualitative risk analysis or numeric rating scales, to describe the potential consequences and likelihood of hazards.

For hazard assessment, the hazard level can be evaluated to ensure that the existing hazards can be prevented, or its consequences can be mitigated. Hazard management will provide suggestions if hazard control measures need to be modified for hazard prevention. Figure 3 shows a technical flow for slope hazard assessment.

2.3. Assessment Methodology Based on ICA

2.3.1. Index Establishment of the Assessment Method

The slope assessment method, which is used to assess highway hazards, is established according to 10 parameters related to the construction scale, geological factors, design, and construction scheme.

Factors that affect slope stability were focused on construction scale and geological factors. Slope height, slope angle, and slope length are three subfactors related to construction scale. The soil layer structure, environmental conditions, groundwater effect and unconfined compressive strength of soil are regarded as four subfactors of geological factors.

External factors include rainfall, earthquakes and human activity during construction. In this paper, the hazard degree during the construction process, stability coefficient of the slope, and design rationality are taken as three subfactors of the external factors for hazard evaluation.

In the evaluation method, ten indices are divided into 3 classifications, which is convenient for qualitative and semiquantitative descriptions (

Table 1. Evaluation method of the index for hazard assessment [26,27].

First-Grade Index	Second-Grade Index and Grading Standard		Score	Illustration
Construction scale (A1, ω1)	Slope height (m) (a11, ω11)	≥45	75–100	These 3 parameters can be calculated based on linear interpolation to obtain the specific values.
		35–45	50–74
		25–35	25–49
		15–25	0–24
	Slope angle (°) (a12, ω12)	>50	75–100
		35–50	50–74
		25–35	25–49
		15–25	0–24
	Slope length (m) (a13, ω13)	≥300	75–100
		200–300	50–74
		100–200	25–49
		<100	0–24
Geological factor (A2, ω2)	Structure of soil layers (a21, ω21)	Loose quaternary alluvial soil	75–100
		Sandstone, mudstone, limestone, shale, and other weathered layers	50–74
		Sandstone, mudstone, limestone and shale, such as full weathered, strongly weathered soil	25–49
		Cataclastic structure, soil–rock mixture	0–24
	Environmental conditions (a22, ω22)	Surrounded by key faults or valleys and rivers	75–100	Field investigation needed, focusing on pipelines, mined out spaces, high-voltage tower, water, etc.
		Close to normal structures	50–74
		No large structure	25–49
		Well geological conditions	0–24
	Groundwater effect (a23, ω23)	Groundwater is exposed at the lower part of slope (<0.25 H), and well water-bearing capability	75–100
		Groundwater is exposed at the lower part of slope (0.25–0.5 H), and well water-bearing capability	50–74
		Groundwater is exposed at the lower part of slope (0.5–0.75 H), and general the water-bearing capability	25–49
		Groundwater is exposed at the lower part of slope (0.75–1.0 H) and the water-bearing is poor	0–24
	Unconfined compressive strength of soil Qu (kpa) (a24, ω24)	0–60	75–100	From the survey design report.
		60–120	50–74
		120–240	25–49
		≥240	0–24
V Design and construction scheme (A3, ω3)	Hazard degree during construction (a31, ω31)	Grouting, slope protection and inclined drainage hole are the main measures, and most of excavation, support and drainage uses labor	75–100	Hazard is judged by the specific engineering measures.
		Anchor, slope protection and inclined drainage hole are the main measures, and some of excavation, support and drainage use labor	50–74
		Retaining wall, drainage channel and slope protection are the main measures, and mechanization is used	25–49
		The slope brushing and slope protection are the main measures, and all work is machine operation	0–24
	Stability coefficient of slope (a32, ω32)	<1.15	75–100	From the survey design report.
		1.15–1.30	50–74
		1.30–1.50	25–49
		≥1.5	0–24
	Design rationality (a33, ω33)	Design data is insufficient; the slope analysis method or parameter calculation is false; the design of water drainage, support system and earthwork is not comprehensive; the construction is not timely	75–100	Design factors affect the construction safety, which should be comprehensively evaluated by experts.
		Design data is insufficient; selected parameters and conditions for construction are insufficient	50–74
		Design qualification is Grade A; employees with experience of 5–10 years can comprehensive understand the design of slope drainage, support and earth excavation	25–49
		Design qualification is Grade A; employees with over 10-years’ experience; timely design and construction	0–24

).

The first-grade index is divided into some specific indices. For example, the construction scale (A₁) is divided into three subgrade indices, including slope height (a₁₁), slope angle (a₁₂), and slope length (a₁₃). All 10 subgrade-specific indices are then divided into 4 grades with corresponding score values from 0 (almost safe) to 100 (very dangerous). The score range of every grade is 25 points.

2.3.2. Proportional Distribution Based on the ICA for the Method

To improve the semiquantitative estimation accuracy, the Index Complexity Algorithm (ICA) was introduced.

The basic idea of the ICA is that if the index is more complex, the amplitude of variation for the specific index is greater. The proportion coefficient reflects the impact degree of the evaluated index during hazard assessment. Therefore, according to the complexity of evaluation factors, a weight distribution is obtained by normalizing the complexity.

The index complexity can be calculated by the following formula:

$$C_j={\displaystyle \frac{2\left|G_{jm2}-G_{j1}-G_{j2}\right|(G_{j2}-G_{j1})}{(G_{jm2}-G_{jm1})}}$$

(1)

where C_j is the complexity of the factor. A higher value reflects a more complicated parameter; conversely, a low value means a relatively simple factor. G_jm2 and G_jm1 are the regional maximum and minimum scores of the factor; usually, the values of G_jm2 and G_jm1 are set as 100 and 0, respectively.

C_j2 and C_j1 are the evaluation data at the specific site.

When the index complexity is obtained, the weight coefficient can be expressed as follows:

$$\omega ={\displaystyle \frac{[C_1,C_2,\dots C_i]}{{\displaystyle \sum C_i}}}$$

(2)

In the evaluation index system, the formula for hazard assessment is expressed as follows:

$$R=A_1\times {\omega }_1+A_2\times {\omega }_2\cdots +A_n\times {\omega }_n$$

(3)

where $A_n=a_{n1}\times {\omega }_{n1}+a_{n2}\times {\omega }_{n2}\cdots +a_{nm}\times {\omega }_{nm}$.

A₁, A₂, …, A_n are the first-grade indices (Column 1 in Table 1). ω₁, ω₂, …, ω_n are the weight values of the first-grade indices (Column 1 of Table 1).

$\sum _{i=1}^n{\omega }_i=1$; a_n1, a_n2, …, a_nm are the second-grade indices of the first grade index A_n. ω_n1, ω_n2, …, ω_nm are the weight values corresponding to the second indices, and $\sum _{i=1}^n{\omega }_{ni}=1$.

Here,

a_nm = (G_nm1 + G_nm2)/2

(4)

During hazard assessment, engineers should conduct a serious geological investigation and become familiar with the geological environmental conditions so that ω_n and ω_nm can be obtained according to the on-site geological information. The detailed procedure will be presented in the case study of the i landslide in the next section.

In addition, based on the results of Formula (3), the hazard levels of high slopes are determined and classified into four levels from low to very high, as shown in

Table 2. Hazard level classification of slope hazard assessment for highway construction.

Hazard Level	Result
Grade IV (Very high)	R > 60
Grade III (High)	40 < R ≤ 60
Grade II (Moderate)	20 < R ≤ 40
Grade I (Low)	R ≤ 20

.

3. Results

3.1. Hazard Analysis and Evaluation

To use the ICA for the assessment indices, 10 indices based on Table 1 are used for the hazard analysis, and then the indices are graded and ranked by considering the specific geological and environmental conditions.

Scores of different indices for the slope were assigned according to the classifications in Table 1:

(1)

Construction scale (A₁): slope height (a₁₁), slope angle (a₁₂), and slope length (a₁₃)

In the assessment method, hazard analysis for construction scale is concentrated on slope height, slope angle, and slope length. Slope height and slope angle usually have a greater impact on stability.

The maximum values for the landslide width and height are approximately 40 m and 120 m. The scores can be calculated by linear interpolation. The complexity for a₁₁ and a₁₃ varies from 50 to 74 and 25 to 49, respectively.

The slope angle is defined as the measured gradient between the cut slope and the road surface. The slope angle varied between 35° and 50° in the slope. The complexity is in the range from 50 to 74.

(2)

Geological conditions (A₂) hazard analysis: soil structure (a₂₁), engineering surrounding structure (a₂₂), groundwater effect (a₂₃), and unconfined compressive strength of soil Qu (kpa) (a₂₄)

The studied landslide is a small-scale, traction-type, mid-layer landslide. The landslide area is covered with silty clay, and the underlying bedrock is the fully weathered layer of the Middle to Upper Silurian Wengxiang Group sandstone interbedded with mudstone, with local areas of strong weathering. The low shear strength of the materials makes the landslide easy to slide with some external factors. Considering the complexity of geological features, the soil structure (a₂₁) ranges from 75 to 80 based on the ICA.

Environmental conditions (a₂₂) are focused on whether there are buildings, buried key linear engineering, high-voltage towers and water facilities near the slope. A V-shaped valley occupies the lower part of the slope. The slope is significantly affected by groundwater and atmospheric rainfall. In the process of rainfall, with the infiltration of rainwater in the slope soil, the transient movement of groundwater, combined with the action of static groundwater, reduces the shear strength of the soil, destroys the stable state of the slope, makes the slope show accelerated damage, and increases the slope deformation. Considering the complexity of geological features, the amplitude for environmental conditions (a₂₂) was assigned in the range from 80 to 95 based on the ICA.

The climate of the slope research area belongs to the middle subtropical monsoon climate, with four distinct seasons, mild climate, abundant precipitation, no severe cold in winter, no hot summer, long frost-free period, rain and heat in the same season, with obvious monsoon climate characteristics. The average annual temperature is 13~16 °C, and the average annual rainfall is 1307.9 mm. Furthermore, 83% of the annual rainfall is concentrated in April to October, with the maximum rainstorm rate in June, and the maximum daily rainfall is 189.9 mm. The annual frost-free period is 282 days, and the severe weather includes heavy rain, spring drought, solar drought, freezing, and hail.

The altitude of the highway is higher than that of the valley. The groundwater is exposed at the lower part of slope with a distance of (0.75–1.0 H). Since the water-bearing is poor and water is largely controlled by rainfall, the variation amplitude of the groundwater effect (a₂₃) was assigned in the range from 0 to 24.

Based on the borehole records, the lithology is silty clay to 7 m and silty clay block with strongly weathered gravel soil to 18 m over pelitic siltstone. The average unconfined compressive strength of soil Qu (kPa) ranges from 118 kPa to 165 kPa. The variation amplitude for the unconfined compressive strength of soil Qu (a₂₄) was assigned in the range from 40 to 55.

(3)

Design and construction scheme (A₃) hazard analysis: hazard degree of construction activity (a₃₁), coefficient of slope stability evaluation (a₃₂), design rationality (a₃₃)

The hazard degree of the construction process (a₃₁) is focused on whether the construction activity exceeds the designed standard. The variation amplitude for the hazard degree of construction activity (a₃₁) was assigned in the range from 75 to 90.

The materials of the slope are very loose and highly permeable, the shear strength of which is poor, causing the landslide to slide easily with triggering factors. The variation amplitude for the stability coefficient of slope evaluation (a₃₂) is assigned in the range from 75 to 85.

The base geological data for design, the selection of parameters, and the various conditions encountered in the construction process are not systematic. The design qualification of the survey team was Grade A. Designers with over five years of engineering experience can comprehensively understand slope drainage, support, and earth excavation. The variation amplitude for design rationality (a₃₃) was then assigned in the range from 65 to 75.

3.2. Semiquantitative Procedure for Slope Hazard Assessment

The following steps for the semiquantitative procedure of slope hazard assessment are outlined on the basis of field geological survey and material analysis.

Step one: Based on the hazard analysis and evaluation for the slope, 10 indices were obtained considering the complexity of the geological and constructional conditions, as shown in Section 3.1. The G_jm2, G_jm1, G_jm2, and G_jm1 values of each evaluation factor can then be further calculated as shown in

Table 3. Scores of the second-grade indices for the construction scale based on G_jm and G_j.

No	Second-Grade Index	Gjm1	Gjm2	Gj1	Gj2
a11	Slope height	0	100	50	74
a12	Slope angle	0	100	50	74
a13	Slope length	0	100	25	49

,

Table 4. Scores of the second-grade index for geological factors based on G_jm and G_j.

No	Second-Grade Index	Gjm1	Gjm2	Gj1	Gj2
a21	Structure of soil layer	0	100	75	80
a22	Environmental conditions	0	100	80	95
a23	Groundwater effect	0	100	0	24
a24	Unconfined compressive strength	0	100	40	55

and

Table 5. Scores of the second-grade index for the design and construction scheme based on G_jm and G_j.

No	Second-Grade Index	Gjm1	Gjm2	Gj1	Gj2
a31	Hazard degree during construction	0	100	75	90
a32	Stability coefficient of slope	0	100	75	85
a33	Design rationality	0	100	65	75

.

Step two: The complexity of every evaluation factor is calculated according to Formula (1) with the data in Table 3, Table 4 and Table 5 as follows:

Complexity for the three indices of construction scale:

[C₁₁, C₁₂, C₁₃] = [11.52, 11.52, 12.48]

Complexity for the four indices of geological factors:

[C₂₁, C₂₂, C₂₃, C₂₄] = [5.5, 22.5, 36.48, 1.5]

Complexity for the three indices of the design and construction scheme:

[C₃₁, C₃₂, C₃₃] = [19.5, 12, 8]

Step three: The weight coefficient can be calculated based on Formula (2).

The weight distributions for the three subindices of the construction scale in Table 3 are ω = [ω₁₁, ω₁₂, ω₁₃] = [0.33, 0.32, 0.35]. Then, the value of A₁ can be obtained by Formulas (3) and (4):

$$\begin{gathered} A_1=a_{11}\times {\omega }_{11}+a_{12}\times {\omega }_{12}+a_{13}\times {\omega }_{13} \\ =49\times 0.33+49\times 0.32+36.5\times 0.35=44.63 \end{gathered}$$

The weight distributions for the four subindices of the geological factor in Table 4 are ω = [ω₂₁, ω₂₂, ω₂₃, ω₂₄] = [0.08, 0.34, 0.56, 0.02]. Then, the value of A₂ can be obtained by Formulas (3) and (4):

$$\begin{gathered} A_2=a_{21}\times {\omega }_{21}+a_{22}\times {\omega }_{22}+a_{23}\times {\omega }_{23}+a_{24}\times {\omega }_{24} \\ =42.5\times 0.08+55\times 0.34+24\times 0.56+35\times 0.02=36.24 \end{gathered}$$

The weight distributions for the three subindices of the design and construction scheme in Table 5 are ω = [ω₃₁, ω₃₂, ω₃₃] = [0.49, 0.3, 0.21]. Then, the value of A₃ can be obtained based on Formulas (3) and (4):

$$\begin{gathered} A_3=a_{31}\times {\omega }_{31}+a_{32}\times {\omega }_{32}+a_{33}\times {\omega }_{33} \\ =52.5\times 0.49+47.5\times 0.3+42.5\times 0.21=48.9 \end{gathered}$$

Step four: Determination of the first-grade evaluation factors.

According to the data in Section 3.1 and Table 3, Table 4 and Table 5, the values of G_jm2, G_jm1, G_j2, and G_j1 for the construction scale, the geological factors, and the construction organization can be obtained based on index complexity (

Table 6. Evaluation of the first-grade indices G_jm and G_j for the slope.

Serial Number	First-Grade Index	Gjm1	Gjm2	Gj1	Gj2
1	Construction scale	0	100	50	90
2	Geological factors	0	100	25	40
3	Construction organization	0	100	65	90

).

The basic idea of the ICA is that if the evaluation index is more complex, the amplitude of variation for the specific index is greater. This reflects the degree of impact of the evaluated index during the hazard assessment. In this case, by multiplying the highest value and lowest value of each second-grade index by their weight coefficients, the corresponding values in columns 5 and 6 of Table 6 can be obtained.

According to Table 6 and Formula (2), the complexity for these three first-grade indices of the landslide are,

[C₁, C₂, C₃] = [32, 30, 27.5].

The primary index weight distribution of the landslide is,

ω = [0.4, 0.3, 0.3].

The scores of the first-grade indices for the landslide are,

[A₁, A₂, A₃] = [44.63, 36.25, 48.9].

According to Formula (3) and the weight values presented above, the score of construction risk assessment for the landslide can be obtained as shown in

Table 7. Construction risk assessment result for the landslide.

Slope	Result
Risk assessment score	43
Risk level	Grade III (High)

, which is 60 points. The risk level of this landslide is Grade III with high risk.

The assessment result suggests that the landslide is very dangerous during excavation for highway construction, and some control measures need to be considered, such as prestressed anchor cables and grouting anchorage. During excavation and reinforcement, dynamic monitoring should be carried out at the same time.

4. Discussion

The risk of landslides is inextricably linked to the probability of landslide damage, and the analysis of landslide damage probability has a certain internal connection with the analysis of landslide stability. Traditional landslide stability analysis is mainly based on the safety factor method, which has a clear calculation principle, is easy to understand, and is adopted by many survey and design units. However, it has certain limitations because it cannot objectively reflect the spatial variations, load effects, and changes in pore water pressure of the research object.

In the risk assessment of geological disasters at home and abroad, the study of natural attributes is mainly emphasized, and the consideration of social attributes is ignored. For example, the research on the impact of construction process on geological disasters is less, and there is a disconnection between the study of natural attributes and social attributes [28]. The indicators are independent of each other, which is obviously divorced from reality. In many existing studies [29,30,31], relevant statistical and mathematical methods (gray system, fuzzy mathematics, hierarchical analysis, neural network, etc.) are selected to evaluate the risk of the slope, often only from the geological conditions of the landslide itself. In the real world, many factors are interrelated and give feedback to each other. The operation ability of the construction units is an important part of it.

As we all know, the calculation results of landslide stability analysis are affected by many external factors, such as the reliability of data sources, the complexity of landslide damage forms, and the limitations of the formula itself, all of which bring a certain degree of uncertainty to the analysis results. It is widely believed that carrying out landslide reliability analysis on the basis of probability theory can obtain more reasonable and reliable results with stronger practicality. The theory of reliability analysis not only fully considers the uncertainty and randomness of various parameters, avoiding the absoluteness and singularity of the calculation results to a great extent, but also allows the probability of damage (i.e., the probability of landslide occurrence) to be obtained on the basis of calculating the landslide safety factor.

To analyze and identify the risk associated with construction, engineers often use analytical techniques such as ETA (event tree analysis), AHP (analytic hierarchy process) or engineering analogue methods to assess the risk of high slopes during construction [16,17,27,32]. However, it is difficult to quantitatively describe construction risk in practical engineering applications.

Table 1 shows that it is impossible to avoid subjective influence during the assessment. The authors thus introduced the ICA method for the weight coefficient distribution in the evaluation system with 10 parameters to reduce the influence of subjectivity during qualitative weight distribution. The basic idea of the ICA is that the more complicated the evaluation index is, the larger the amplitude of variation for the specific index. Ten indices in the assessment method are relatively comprehensive and representative, which describes the important impact on construction safety.

During assessment, evaluation of G_jm2, G_jm1, G_j2, and G_j1 should be based on the specific geological and environmental conditions of the site. For example, when part of the slope is under the groundwater level or immersed in the water of the river/lake, the strength of matter will be greatly weakened; some structural planes, such as the potential sliding face in the slope, have the most prominent influence on slope stability.

The landslide susceptibility assessment method (ICA method) has been officially adopted by the Ministry of Transport of China and incorporated into the “Highway Cutting High Slope Engineering Construction Safety Risk Assessment Guide”. The guide mandates that all newly built, reconstructed, or expanded highway projects included in the national and local basic construction plans must apply this assessment method during the construction phase to ensure safety and quality of the project.

In the past few years, the assessment method developed by the authors has been applied in many provinces and cities of China and has achieved good results. It has received increasing attention from Chinese researchers and engineers. It is still in the process of continuous improvement.

This study’s proposed landslide susceptibility assessment method not only enhances the objectivity and precision of the evaluation but also contributes to the long-term stability and environmental sustainability of mountain highways by reducing the potential for landslide-induced damage to infrastructure. The application of this method helps to minimize future repair work, conserve resources, and reduce environmental disruption, all of which are key factors in achieving sustainable development.

5. Conclusions

This paper introduces an Index Complexity Algorithm (ICA) and establishes a comprehensive evaluation method for the stability of slopes during construction. This is a hierarchical evaluation method with good scientificalness and simplicity. The evaluation indices in the assessment method were divided into two grades with three categories. Most of the indices were selected according to geological investigation. The index system (Table 1) uses 0–100 to assign scores during evaluation. Among them, the construction scale, geological factor, design, and construction scheme are relatively easy to collect in practical assessment.

The following conclusions have been drawn:

(1)

This paper fully considers both subjective and objective factors affecting landslide occurrence and establishes an evaluation method for highway construction slopes, which includes 10 parameters related to construction scale, geological factors, and engineering design.

(2)

The newly proposed Index Complexity Algorithm (ICA) overcomes the shortcomings of traditional evaluation methods where subjective judgments affect the weight of factors. The new method can profoundly reflect the differentiation of factors, objectively determine the weights of evaluation factors, and is simple and fast to operate.

(3)

The evaluation method proposed in this paper still has certain limitations, such as the inability to evaluate stability under dynamic conditions that change over time (e.g., rainfall). It is more suitable for the evaluation of landslide stability under current conditions. Future surveys and research should refine this method to better apply it to the stability evaluation of landslides under various working conditions.

In summary, the assessment method introduced in this paper makes a significant contribution to promoting sustainable development by ensuring the safety and stability of mountain highway infrastructure through effective landslide risk management. Future research will continue to refine this method to better adapt to changing environmental conditions and further support the development of sustainable transportation infrastructure.