OPEN ACCESS
In urban rail transit projects, the traditional method of foundation evaluation faces problems like vague description, unclear process, and fuzzy evaluation system. To solve these problems, this paper sets up a scientific evaluation system based on CGB technology integration and analytic hierarchy process (AHP). The CGB technology integration refers to the integrated application of computeraided design (CAD), geographic information system (GIS), and building information modeling (BIM). Taking Qingxihe Station, Line 6 of Chongqing Rail Transit (CRT) as the object, the authors constructed a 3D geological model of the construction site, created a novel threelayer system through data analysis, and evaluated and compared the suitability of each layer as the supporting layer of the foundation. Finally, effective suggestions were put forward on the selection of the supporting layer. Our research successfully visualizes the whole process of foundation evaluation, and enhances the accuracy of the evaluation results. The research findings provide a good reference for the selection of the supporting layer of foundations in urban rail transit projects.
CGB technology integration, foundation evaluation, analytic hierarchy process (AHP), visualization
The CGB technology integration refers to the integrated application of computeraided design (CAD), geographic information system (GIS), and building information modeling (BIM). To meet the various needs of urban digitalization, the CGB technology integration jointly utilizes macrogeographic information and microbuilding information, facilitating queries and analyses [1].
Compared with the CAD data, the BIM data and GIS data are not highly compatible. To share and merge these data, it is necessary to explore the industry foundation classes (IFC) model of the BIM and the CityGML model of the GIS, and develop a method capable of automatically extracting the GIS surface model with multiple levels of details (LODs) from the BIM entity model. In this way, the LOD 100400 models could be obtained from the IFC and CityGML model, overcoming the difficulty in merging the BIM with the GIS and paving the way for the CGB technology integration [214].
The BIM and GIS are the two most popular digital technologies in the research of urban rail transit. For instance, D’Amico et al. [15] integrated the BIM with the GIS into the design of transport infrastructure, and suggested that the interoperable sharing models supplemented by the GIS data could minimize or eliminate the possible conflicts between infrastructure design and environmental constraints. Liu et al. [16] fully utilized the advantages of the BIM (e.g. 3D visualization, parametrization, and virtual simulation) to solve foundation engineering, a basic problem in rail transit, and thus improved the quality and efficiency of metro construction. Chen et al. [17] realized the conversion between geometric and semantic information through the BIM and 3D GIS data exchange method of rail transit, and defined an integrated 3D spatial data model, achieving unified management and seamless expression of the data on rail transit and its surroundings. He et al. [18] displayed and analyzed the spatial distribution of unfavorable geological bodies in the GIS, evaluated the karst collapse risk in the area crossed by the tunnel, and assessed the safety risk of metro tunnel on the BIM platform based on construction and monitoring information, laying the basis for tunnel safety prewarning.
Based on CGB technology integration and analytic hierarchy process (AHP), this paper develops a scientific method to visualize the foundation evaluation in urban rail transit projects. Qingxihe Station, Line 6 of Chongqing Rail Transit (CRT) was taken as the research case to evaluate the suitability of different layers to serve as the support layer of the foundation. The authors detailed the selection of evaluation method and the establishment of evaluation system, and verified the proposed method through case analysis, shedding new light on the visualization of the results of engineering geological investigation.
In engineering geological investigation, engineering geological evaluation is one of the key contents. The utmost goal of engineering geological investigation lies in foundation evaluation, a part of engineering geological evaluation. Through foundation evaluation, the suitability of each layer as the supporting layer of the foundation could be quantified. The CGB technology integration provides a desirable tool to visualize the foundation evaluation. As shown in Figure 1, the CGBbased visualization of foundation evaluation mainly acquires the spatial distribution of the geological information in the construction site through investigation and field survey, models the spatial situation with discrete data points, and then analyzes the geological data using the information retrieval and processing functions of CGB technology integration [1926].
Figure 1. The roadmap of CGBbased visualization of foundation evaluation
2.1 Selection of evaluation method
According to the response from construction parties, there are several problems with the current method for foundation evaluation: the foundation quality is described vaguely by qualitative words (e.g. general, good, and poor), without any quantitative comparison; the evaluation items are complex and not weighted; the evaluation process and results derivation are not visible. To solve these problems, the AHP was introduced to provide a multifactor evaluation system for foundation evaluation, and fully integrate qualitative analysis with quantitative analysis.
2.2 Establishment of evaluation system
(1) Level division
The contents of foundation evaluation were divided as per the requirements of relevant codes. According to the goal, items, and objects of foundation evaluation, a multilayer evaluation model was established, in which each layer controls and is controlled by its upper and lower layers. As shown in Figure 2, the established model consists of the goal layer (evaluation goal), the criteria layer (evaluation items), and the alternative layer (evaluation objects).
(2) Construction of judgment matrix for pairwise comparison
Once the evaluation model is established, it is necessary to determine the judgment matrix of each layer, that is, to judge the relative importance of each factor on each layer and express it as a numerical value. Based on the hierarchy of the three layers, it is also necessary to determine the importance of each factor on the lower layer relative to each relevant factor on the upper layer (goal A or criterion Z). Suppose factor A_{k} on layer A is correlated with factors B_{1}, B_{2}, …, B_{n} on the lower layer. Then, the judgement matrix of layer A can be constructed as Table 1, where A_{k} is the numerical value of the importance of B_{i} relative to B_{j}. The relative importance is usually rated against a ninepoint scale (Table 2).
Table 1. The judgement matrix
A_{k} 
B_{1} 
B_{2} 
… 
B_{n} 
B_{1} B_{2} $\vdots$ B_{n} 
b_{11} b_{21} $\vdots$ b_{n1} 
b_{12} b_{22} $\vdots$ b_{n2} 
… … $\vdots$ … 
b_{1n} b_{2n} $\vdots$ b_{nn} 
Levels 
Meanings 
1 
Two factors are equally important. 
3 
The former factor is slightly more important than the latter. 
5 
The former factor is strongly more important than the latter. 
7 
The former factor is very strongly more important than the latter. 
9 
The former factor is extremely more important than the latter. 
2, 4, 6, 8 
The relative importance falls between two of the above levels. 
Reciprocals of above 
If the importance of factor i relative to factor j is b_{ij}, then the importance of factor j relative to factor i is b_{ji}=1/b_{ij}. 
The judgment matrix must also satisfy:
$b_{i j}>0 ; b_{j i}=1 / b_{i j} ; b_{i i}=1(i=1,2, \cdots, n)$ (1)
Formula (1) shows that the judgement matrix is symmetric. In special cases, the judgment matrix must also be transitive:
${{b}_{ij}}\cdot {{b}_{jk}}={{b}_{ik}}$ (2)
(3) Single ranking
Single ranking is to sort the factors on a layer by the importance relative to each relevant factor on the upper layer. The single ranking is equivalent to the calculation of the characteristic roots and eigenvectors of the judgement matrix. In other words, judgement matrix B should satisfy:
$BW={{\lambda }_{\max }}W$ (3)
where, $\lambda_{\max }$ is the maximum characteristic root of B; W is the normalized eigenvector corresponding to $\lambda_{\max }$; W_{i}, a component of W, is the weight of factor i in single ranking.
Besides, the consistency of the judgment matrix should be verified by computing its consistency index CI:
$CI=\frac{{{\lambda }_{\max }}n}{n1}$ (4)
If the judgement matrix is fully consistent, CI=1; the greater the $\lambda_{\max }n$, the larger the CI, and the less consistent is the judgement matrix. Since the sum of n eigenvalues of B equals n, CI is equivalent to the mean of n1 characteristic roots other than $\lambda_{\max }$.
Table 3. The mean random consistency index (RI)
Order 
1 
2 
3 
4 
5 
6 
7 
8 
9 
RI 
0.00 
0.00 
0.58 
0.90 
1.12 
1.24 
1.32 
1.41 
1.45 
The consistency of overall ranking results should be verified in a similar manner. From top to bottom, the consistency needs to be checked layer by layer. Let CI_{j}^{(k)} and RI_{j}^{(k)} be the CI and RI of a factor on layer k relative to factor j on layer k1 in single ranking, respectively. Then, the CR of layer k in overall ranking can be expressed as:
$C{{R}^{(\text{k})}}=\frac{\sum\limits_{j1}^{{{n}_{i}}}{w_{j}^{(k1)}CI_{j}^{(k)}}}{\sum\limits_{j1}^{{{n}_{i}}}{w_{j}^{(k1)}RI_{j}^{(k)}}}$ (5)
Similarly, if CR^{(k)}≤0.10, the overall ranking results have satisfactory consistency.
(4) Overall ranking
Overall ranking is to sort the importance of all factors on the current layer relative to the superior layer, based on the single ranking results of the current layer relative to all the other layers. The overall ranking needs to be performed layer by layer from top to bottom. Suppose the importance ranking of the n factors on layer k1 relative to the goal layer satisfy:
${{w}^{(k1)}}={{(w_{1}^{(k1)},\cdots ,w_{n}^{(k1)})}^{T}}$ (6)
The single ranking vector of n_{k} factors on layer k relative to criterion j on layer k1 can be defined as:
$\text{u}_{\text{j}}^{(k)}={{(\text{u}_{\text{1j}}^{(k)},\text{u}_{\text{2j}}^{(k)},\cdots ,\text{u}_{{{\text{n}}_{\text{i}}}\text{j}}^{(k)})}^{T}}$. j=1, 2, ···, n; k=1, 2, ···, n_{k} (7)
The importance of factors not linked to criterion j was set to zero. Then, a matrix of order n_{k}×n can be obtained:
${{U}^{(k)}}=\text{(u}_{\text{1}}^{\text{(k)}}\text{,u}_{\text{2}}^{\text{(k)}}\text{,}\cdots \text{,u}_{\text{n}}^{\text{(k)}}\text{)}=\left( \begin{matrix} \text{u}_{\text{11}}^{\text{(k)}} & \text{u}_{\text{12}}^{\text{(k)}} & \cdots & \text{u}_{\text{1n}}^{\text{(k)}} \\ \text{u}_{\text{21}}^{\text{(k)}} & \text{u}_{\text{22}}^{\text{(k)}} & \cdots & \text{u}_{\text{2n}}^{\text{(k)}} \\ \vdots & \vdots & \vdots & \vdots \\ \text{u}_{{{\text{n}}_{\text{i}}}\text{1}}^{\text{(k)}} & \text{u}_{{{\text{n}}_{\text{i}}}2}^{\text{(k)}} & \cdots & \text{u}_{{{\text{n}}_{\text{i}}}\text{n}}^{\text{(k)}} \\\end{matrix} \right)$ (8)
where, column j in U^{(k)} is the single ranking vector of n_{k} factors on layer k relative to criterion j on layer k1. Then, the overall ranking of all factors on layer k can be expressed as:
${{w}^{(k)}}={{(w_{1}^{(k)},\cdots ,w_{n}^{(k)})}^{T}}$ (9)
Then,
$\begin{align} & {{\text{w}}^{(k)}}={{U}^{(k)}}{{\text{w}}^{(k1)}} =\left( \begin{matrix} \text{u}_{\text{11}}^{\text{(k)}} & \text{u}_{\text{12}}^{\text{(k)}} & \cdots & \text{u}_{\text{1n}}^{\text{(k)}} \\ \text{u}_{\text{21}}^{\text{(k)}} & \text{u}_{\text{22}}^{\text{(k)}} & \cdots & \text{u}_{\text{2n}}^{\text{(k)}} \\ \vdots & \vdots & \vdots & \vdots \\ \text{u}_{{{\text{n}}_{\text{i}}}\text{1}}^{\text{(k)}} & \text{u}_{{{\text{n}}_{\text{i}}}2}^{\text{(k)}} & \cdots & \text{u}_{{{\text{n}}_{\text{i}}}\text{n}}^{\text{(k)}} \\\end{matrix} \right)\left( \begin{matrix} \text{w}_{\text{1}}^{\text{(k1)}} \\ \text{w}_{\text{2}}^{\text{(k1)}} \\ \vdots \\ \text{w}_{\text{n}}^{\text{(k1)}} \\\end{matrix} \right)=\left( \begin{matrix} \sum\limits_{j1}^{n}{\text{u}_{\text{1j}}^{\text{(k)}}\text{w}_{\text{j}}^{\text{(k1)}}} \\ \sum\limits_{j1}^{n}{\text{u}_{\text{2j}}^{\text{(k)}}\text{w}_{\text{j}}^{\text{(k1)}}} \\ \vdots \\ \sum\limits_{j1}^{n}{\text{u}_{{{\text{n}}_{\text{i}}}\text{j}}^{\text{(k)}}\text{w}_{\text{j}}^{\text{(k1)}}} \\\end{matrix} \right) \\ \end{align}$ (10)
That is,
$\text{w}_{i}^{(k)}=\sum\limits_{\text{j1}}^{\text{n}}{u_{ij}^{(k)}}w_{j}^{(k1)}$, i=1, 2, ···, n (11)
Through the above steps, the score of each alternative (evaluation object) can be obtained. The score ranking determines the relative importance (suitability) of each object to the goal.
3.1 Project overview
The case project is Qingxihe Station, Line 6 of CRT. Lying below Yuegang Longitudinal Road and Yuegang Middle Road, the northsouth oriented station crosses the interaction between the two roads, adopts an open cut doublelayer rectangular frame, and has a 12mlong island platform. The total length, maximum clear width, and maximum clear height are 290.9m, 24.3m, and 16.31m, respectively. There are two air ducts and six entrances and exits (two of which are reserved), all of which adopt opencut rectangular frames.
3.2 CGBbased foundation evaluation
The main construction project of case is a metro station (an underground space project) and its ancillary works. The following items should be considered to evaluate the foundation of the project: uniformity of each layer, sorting of overburden composition, thickness of each layer, mechanical properties of each layer, groundwater effect, and adverse geological phenomena.
3.2.1 Establishment of evaluation model
Based on the above items, an AHP structure was set up, consisting of a goal layer, a criteria layer, and an alternative layer (as shown in Figure 3).
According to the complexity classification of geological environment in relevant codes, the importance of evaluation items in the project were ranked as mechanical properties, adverse geological phenomena, groundwater effect, uniformity, thickness, and sorting. On this basis, the judgement matrix of the criteria layer was established as Table 4.
The consistency of the judgement matrix was computed as 0.0156, indicating that the matrix is sufficiently consistent.
Figure 3. The AHP structure
Table 4. The judgement matrix of the criteria layer

Uniformity 
Sorting 
Thickness 
Mechanical properties 
Groundwater depth 
Groundwater seasonality 
Adverse geological phenomena 
Uniformity 

3 
2 
1/7 
1/3 
1/3 
1/5 
Sorting 


1/2 
1/9 
1/5 
1/5 
1/7 
Thickness 



1/8 
1/4 
1/4 
1/6 
Mechanical properties 




3 
3 
2 
Groundwater depth 





1 
1/2 
Groundwater seasonality 






1/2 
Adverse geological phenomena 







After setting up the judgement matrix of the criteria layer, it is necessary to establish a judgement matrix of each item for plain fill, silty clay, strongly weathered sandy mudstone, moderately weathered sandstone, and moderately weathered sandy mudstone.
(1) Uniformity evaluation
Taking silty clay for example, the 3D model data of uniformity were imported to ArcScene. Then, the 3D Analyst tool was called from the ArcToolBox to convert the triangulated irregular network (TIN) model (as shown in Figure 4) of the upper and lower surfaces of the silty clay layer into grids. Then, a new grid map (as shown in Figure 5) was obtained by removing the grids of the upper and lower surfaces, revealing the thickness of silty clay across the construction site.
Next, the classification parameters were configured to reclassify the grids, producing a bar chart on the thickness of silty clay across the construction site (as shown in Figure 6). The bar chart visually displays the thickness data of silty clay in any location of the site. Then, the uniformity of the silty clay layer was judged by the standard deviation (SD) and the proportion of the thickness concentration area in the total area. The uniformities of the other layers were obtained in a similar manner. The uniformities of all layers are summed up in Table 5.
The plain fill is sporadically distributed in the construction site, showing a poor uniformity. Based on Table 5, the layers could be ranked in descending order of uniformity: silty clay, moderately weathered sandy mudstone, moderately weathered sandstone, strongly weathered sandy mudstone, and plain fill. On this basis, the judgement matrix of uniformity was established as Table 6.
Figure 4. The TIN model of geological information
Figure 5. The grid map
Figure 6. The bar chart of silty clay thickness
(2) Sorting evaluation
The sorting of each layer against the overburden directly affects the bearing capacity of the foundation, exerting a huge impact on the stability of buildings on the surface. The sorting quality mainly depends on the uniformity of the size of clastic particles. The more uniform the size, the better the sorting.
According to the project data, the plain fill is mainly composed of sandstone and sandy mudstone blocks (fragments). The size of the skeleton particles falls within 20500mm, and could surpass 1m in local areas. The content of these particles is generally 2030%. In relatively thick sections, the content of these blocks (fragments) increases significantly to 7080% in local areas, while the particle size also increases. In the plain fill, the skeleton particles have a nonuniform distribution of content, and significant changes in particle size. By contrast, there is no obvious inclusion in silty clay. This layer mainly consists of clay, with few hard matters.
Through the above analysis, the layers could be ranked in descending order of sorting: moderately weathered sandy mudstone/moderately weathered sandstone, strongly weathered sandy mudstone, silty clay, and plain fill. On this basis, the judgement matrix of sorting was established as Table 7.
(3) Thickness evaluation
For construction projects, the supporting layer cannot be stable without a good and uniform distribution. Besides, the thickness of the supporting layer could affect the basic design of the foundation.
To clearly understand the thickness of layers beneath the construction site, the CGB technology integration was introduced to quantify the thickness of each layer through ArcGIS data analysis, and display the analysis results on 3D models. The mean and proportion of thickness concentration area in Table 5 were referenced to evaluate the thickness of each layer.
Through the above analysis, the layers could be ranked in descending order of thickness: moderately weathered sandy mudstone, silty clay, strongly weathered sandy mudstone, moderately weathered sandstone, and plain fill. On this basis, the judgement matrix of thickness was established as Table 8.
(4) Evaluation of mechanical properties
The mechanical properties of soilrock mass must be considered in foundation evaluation and foundation design. Whether a layer is suitable as the supporting layer largely depends on the quality of its mechanical parameters.
The mechanical properties of each layer were evaluated against the standard bearing capacity of rock foundation mentioned in the project data. Then, the layers were ranked in descending order of mechanical properties as Table 9, where the figures marked with an asterisk are derived from relevant codes and empirical values of the region.
As shown in Table 9, the layers could be ranked in descending order of mechanical properties: moderately weathered sandstone, moderately weathered sandy mudstone, strongly weathered sandy mudstone, silty clay, and plain fill. On this basis, the judgement matrix of mechanical properties was established as Table 10.
Table 5. The uniformities of all layers
No. 
Layers 
Min. (m) 
Max. (m) 
Mean (m) 
SD 
Thickness concentration area (m) 
Proportion of thickness concentration area (%) 
1 
Silty clay 
7.52 
19.94 
12.95 
2.49 
1015 
75 
2 
Strongly weathered sandy mudstone 
5.76 
25.40 
7.38 
4.64 
510 
45 
3 
Moderately weathered sandstone 
0 
26.92 
5.22 
4.88 
05 
51 
4 
Moderately weathered sandy mudstone 
1.12 
13.40 
2.94 
3.38 
05 
57 

Plain fill 
Silty clay 
Strongly weathered sandy mudstone 
Moderately weathered sandstone 
Moderately weathered sandy mudstone 
Plain fill 

1/9 
1/3 
1/5 
1/7 
Silty clay 


4 
3 
2 
Strongly weathered sandy mudstone 



1/2 
1/4 
Moderately weathered sandstone 




1/2 
Moderately weathered sandy mudstone 






Plain fill 
Silty clay 
Strongly weathered sandy mudstone 
Moderately weathered sandstone 
Moderately weathered sandy mudstone 
Plain fill 

1/3 
1/8 
1/9 
1/9 
Silty clay 


1/7 
1/8 
1/8 
Strongly weathered sandy mudstone 



1/2 
1/2 
Moderately weathered sandstone 




1 
Moderately weathered sandy mudstone 






Plain fill 
Silty clay 
Strongly weathered sandy mudstone 
Moderately weathered sandstone 
Moderately weathered sandy mudstone 
Plain fill 

1/7 
1/5 
1/3 
1/9 
Silty clay 


3 
2 
1/3 
Strongly weathered sandy mudstone 



2 
1/2 
Moderately weathered sandstone 




1/3 
Moderately weathered sandy mudstone 





Layers Type of index 
Silty clay 
Plain fill 
Moderately weathered sandstone 
Strongly weathered sandy mudstone 
Moderately weathered sandy mudstone 
Standard bearing capacity of rock foundation (kPa) 
130* 
100* 
1200 
350* 
850 

Plain fill 
Silty clay 
Strongly weathered sandy mudstone 
Moderately weathered sandstone 
Moderately weathered sandy mudstone 
Plain fill 

1/2 
1/4 
1/9 
1/8 
Silty clay 


1/2 
1/9 
1/8 
Strongly weathered sandy mudstone 



1/4 
1/3 
Moderately weathered sandstone 




2 
Moderately weathered sandy mudstone 





Groundwater effect was included in foundation evaluation, because most structures of the project are underground and affected by groundwater. The spatial information of survey points was imported to ArcGIS, and then projected to the geological model of the site, producing a 3D model of groundwater depth (as shown in Figure 7).
Based on project data and the empirical values of the region, the permeability coefficients of plain fill, silty clay, strongly weathered sandy mudstone, moderately weathered sandstone, and moderately weathered sandy mudstone were obtained as 5×10^{5}cm/s, 5×10^{6}cm/s, 2×10^{5}cm/s, 1.2×10^{5}cm/s, and 2×10^{6}cm/s, respectively.
Through the above analysis, the layers could be ranked in descending order of groundwater effect: moderately weathered sandy mudstone, silty clay, moderately weathered sandstone, strongly weathered sandy mudstone, and plain fill. On this basis, the judgement matrix of groundwater effect was established as Table 11.
(6) Adverse geological phenomena evaluation
Adverse geological phenomena usually refer to the geological phenomena in and around the construction site that are not conducive to engineering construction, such as landslides, debris flows, ground collapses, and hidden karsts.
The project data show that the construction site, located on the east wing of the Yuelai syncline, has a normal stratigraphic sequence, without adverse geological effects like landslides, ground collapses, or faults. Hence, the factors in the judgement matrix of adverse geological phenomena are of equal importance.
3.2.3 Results of foundation evaluation
Based on the evaluation system and the judgment matrix of the criteria layer, the suitability of each layer as supporting layer of the foundation was calculated according to the results of each judgement matrix, yielding the results of foundation evaluation.
(1) Display of weight distribution
Through single ranking and overall ranking, the weights of all criteria of our model were obtained (as shown in Figure 8).
(2) Results of the judgement matrix of the criteria layer
The results of the judgement matrix of the criteria layer are presented in Table 12 below.
(3) Evaluation results and comparison chart
The criteria weights were coupled with the results of the judgement matrix of the criteria layer to produce the suitability of different layers (Table 13, Figure 9). Obviously, moderately weathered sandstone and moderately weathered sandy mudstone are the suitable supporting layers of the foundation.
3.2.4 Results analysis
In the survey report of the metro station, there is a complete section of foundation evaluation. This section reports that, the construction site is distributed with silty clay and a small amount of plain fill, sandstone, and sandy mudstone, as confirmed by drilling, geological mapping, and survey. According to the engineering geological features of soilrock mass in the site, the upper fill cannot serve as the supporting layer of the foundation, due to the great variation in thickness, poor uniformity, and low strength. The silty clay in the lower part will cause differential settlement, if it acts as the supporting layer: the silty clay varies greatly in thickness, and exists as lenticles or thins out in local areas, not to mention its poor strength. However, the moderately weathered rocks in the lower part are an ideal supporting layer of the foundation, thanks to their high strength and stability. The conclusion of the report agrees well with our AHP results.
Figure 7. The model of groundwater depth
Table 11. The judgement matrix of groundwater effect

Plain fill 
Silty clay 
Strongly weathered sandy mudstone 
Moderately weathered sandstone 
Moderately weathered sandy mudstone 
Plain fill 

1/7 
1/3 
1/5 
1/9 
Silty clay 


1/2 
2 
1/3 
Strongly weathered sandy mudstone 



1/2 
1/4 
Moderately weathered sandstone 




1/2 
Moderately weathered sandy mudstone 





Items 
Characteristic roots 
Eigenvectors 
CR values 
Bar charts 
Uniformity 
5.0691 
(0.037, 0.4284, 0.0907, 0.16, 0.284)^{T} 
0.0154 
(a) The bar chart of uniformity of different layers 
Sorting 
5.1700 
(0.0293, 0.0496, 0.2231, 0.349, 0.349)^{T} 
0.0379 
(b) The bar chart of sorting of different layers 
Thickness 
5.2260 
(0.0366, 0.2642, 0.1663, 0.1103, 0.4225)^{T} 
0.0504 
(c) The bar chart of thickness of different layers 
Mechanical properties 
5.0962 
(0.0364, 0.0546, 0.1171, 0.471, 0.321)^{T} 
0.0215 
(d) The bar chart of mechanical properties of different layers 
Groundwater depth 
5.0000 
(0.2,0.2,0.2,0.2, 0.2)^{T} 
0 
(e) The bar chart of groundwater depth of different layers 
Groundwater seasonality 
5.4231 
(0.0371, 0.1919, 0.157, 0.1833, 0.4306)^{T} 
0.0944 
(f) The bar chart of groundwater seasonality of different layers 
Adverse geological phenomena 
5.0000 
(0.2,0.2,0.2,0.2, 0.2)^{T} 
0 
(g)The bar chart of adverse geological phenomena of different layers 
Layers 
Plain fill 
Silty clay 
Strongly weathered sandy mudstone 
Moderately weathered sandstone 
Moderately weathered sandy mudstone 
Suitability 
0.0973 
0.1569 
0.1567 
0.2955 
0.2935 
Figure 8. The pie chart of weight distribution
Figure 9. The suitability of each layer
This paper successfully visualizes foundation evaluation based on CGB technology integration and the AHP. Multiple influencing factors were quantified, making the evaluation process and results more reliable. The evaluation process is completely visible, and the evaluation results are highly accurate, allowing every construction party to check the judgement of each factor easily and independently. The research findings provide a good reference for the selection of the supporting layer of foundations in urban rail transit projects.
This work was supported by Chongqing Social Career and People’s Livelihood Guarantee Science and Technology Innovation Special Program (cstc2016shmszx30017); Chongqing Fundamental and Frontier Research (cstc2017jcyjAX0260); Chongqing Graduate Education Innovation Fund (CYS18227).
[1] Wang, H., Pan, Y., Luo, X. (2019). Integration of BIM and GIS in sustainable built environment: A review and bibliometric analysis. Automation in Construction, 103: 4152. https://doi.org/10.1016/j.autcon.2019.03.005
[2] Colucci, E., De Ruvo, V., Lingua, A., Matrone, F., Rizzo, G. (2020). HBIMGIS integration: From IFC to cityGML standard for damaged cultural heritage in a multiscale 3D GIS. Applied Sciences, 10(4): 1356. https://doi.org/10.3390/app10041356
[3] Stouffs, R., Tauscher, H., Biljecki, F. (2018). Achieving complete and nearlossless conversion from IFC to CityGML. ISPRS International Journal of GeoInformation, 7(9): 355163. https://doi.org/10.3390/ijgi7090355
[4] Jusuf, S.K., Mousseau, B., Godfroid, G., Soh, J.H.V. (2017). Path to an integrated modelling between IFC and CityGML for neighborhood scale modelling. Urban Science, 1(3): 25. https://doi.org/10.3390/urbansci1030025
[5] Deng, Y., Cheng, J. C., Anumba, C. (2016). Mapping between BIM and 3D GIS in different levels of detail using schema mediation and instance comparison. Automation in Construction, 67: 121. https://doi.org/10.1016/j.autcon.2016.03.006
[6] Diakite, A.A., Zlatanova, S. (2020). Automatic georeferencing of BIM in GIS environments using building footprints. Computers, Environment and Urban Systems, 80: 101453. https://doi.org/10.1016/j.compenvurbsys.2019.101453
[7] Tsilimantou, E., Delegou, E.T., Nikitakos, I.A., Ioannidis, C., Moropoulou, A. (2020). GIS and BIM as integrated digital environments for modeling and monitoring of historic buildings. Applied Sciences, 10(3): 1078. https://doi.org/10.3390/app10031078
[8] Hijazi, I., Donaubauer, A., Kolbe, T.H. (2018). BIMGIS integration as dedicated and independent course for geoinformatics students: Merits, challenges, and ways forward. ISPRS International Journal of GeoInformation, 7(8): 319. https://doi.org/10.3390/ijgi7080319
[9] Koutamanis, A. (2020). Dimensionality in BIM: Why BIM cannot have more than four dimensions? Automation in Construction, 114: 103153. https://doi.org/10.1016/j.autcon.2020.103153
[10] Kumar, K., Labetski, A., Ohori, K.A., Ledoux, H., Stoter, J. (2019). The LandInfra standard and its role in solving the BIMGIS quagmire. Open Geospatial Data, Software and Standards, 4(1): 116. https://doi.org/10.1186/s409650190065z
[11] Arroyo Ohori, K., Diakité, A., Krijnen, T., Ledoux, H., Stoter, J. (2018). Processing BIM and GIS models in practice: experiences and recommendations from a GeoBIM project in the Netherlands. ISPRS International Journal of GeoInformation, 7(8): 311318. https://doi.org/10.3390/ijgi7080311
[12] Zhang, L., ElGohary, N.M. (2020). Automated IFCbased building information modelling and extraction for supporting value analysis of buildings. International Journal of Construction Management, 20(4): 269288. https://doi.org/10.1080/15623599.2018.1484850
[13] Mirarchi, C., Pavan, A., De Marco, F., Wang, X., Song, Y. (2018). Supporting facility management processes through endusers’ integration and coordinated BIMGIS technologies. ISPRS International Journal of GeoInformation, 7(5): 191197. https://doi.org/10.3390/ijgi7050191
[14] Marzouk, M., Othman, A. (2020). Planning utility infrastructure requirements for smart cities using the integration between BIM and GIS. Sustainable Cities and Society, 57: 120126. https://doi.org/10.1016/j.scs.2020.102120
[15] D’Amico, F., Calvi, A., Schiattarella, E., Di Prete, M., Veraldi, V. (2020). BIM and GIS data integration: A novel approach of technical/environmental decisionmaking process in transport infrastructure design. Transportation Research Procedia, 45: 803810. https://doi.org/10.1016/j.trpro.2020.02.090
[16] Liu, B., Sun, X. (2018, March). Application analysis of BIM technology in metro rail transit. In IOP Conference Series: Earth and Environmental Science, 128: 2831. https://doi.org/10.1088/17551315/128/1/012028
[17] Chen, G., Xue, M., Hu, Z.J., Liu, Y.Z. (2019). GIS+BIM spatial framework for urban rail transit project. Bulletin of Surveying and Mapping, (S2): 262266. https://doi.org/10.13474/j.cnki.112246.2019.0639
[18] He, G.F., Luo, X.Q., Zhang, H. (2019). Technology of early warning and forecast of metro tunnel safety based on BIM and GIS. Urban Mass Transit, (7): 161164. https://doi.org/10.16037/j.1007869x.2019.07.038
[19] Dzikunoo, E.A., Vignoli, G., Jørgensen, F., Yidana, S. M., BanoengYakubo, B. (2020). New regional stratigraphic insights from a 3D geological model of the Nasia subbasin, Ghana, developed for hydrogeological purposes and based on reprocessed Bfield data originally collected for mineral exploration. Solid Earth, 11(2): 349361. https://doi.org/10.5194/se113492020
[20] Chen, L., Wang, H., Xu, X., Zhang, Y., Wang, C., Song, J., Han, L. (2019). Geological exploration using integrated geophysical methods in tunnel: A case. Geotechnical and Geological Engineering: An International Journal, 38(2): 11111119. https://doi.org/10.1007/s1070601901075w
[21] Ku, T., Palanidoss, S., Zhang, Y., Moon, S.W., Wei, X., Huang, E.S., Goh, K.H. (2020). Practical configured microtremor array measurements (MAMs) for the geological investigation of underground space. Underground Space, 19: 112. https://doi.org/10.1016/j.undsp.2020.01.004
[22] Kanik, M., Ersoy, H. (2019). Evaluation of the engineering geological investigation of the Ayvali dam site (NE Turkey). Arabian Journal of Geosciences, 12(3): 89. https://doi.org/10.1007/s1251701942431
[23] Soldo, L., Vendramini, M., Eusebio, A. (2019). Tunnels design and geological studies. Tunnelling and Underground Space Technology, 84, 8298. https://doi.org/10.1016/j.tust.2018.10.013
[24] Di Giulio, G., Ercoli, M., Vassallo, M., Porreca, M. (2020). Investigation of the Norcia basin (Central Italy) through ambient vibration measurements and geological surveys. Engineering Geology, 267: 105501. https://doi.org/10.1016/j.enggeo.2020.105501
[25] Zhang, C.C., Gu, P., Cao, F.X. (2020). Discussion on the hazards of hydrogeological problems in engineering geological survey. Construction & Design for Engineering, 4041. https://doi.org/10.13616/j.cnki.gcjsysj.2020.03.220
[26] Gao, Y., Liang, G.H., Zhou, Y.Y. (2019). Exploration of geotechnical engineering investigation under complex topographical and geological conditions. Frontiers Research of Architecture and Engineering, 2(4): 2023. https://doi.org/10.30564/frae.v2i4.1512