IoT Monitoring and Evaluating System for the Construction Quality of Foundation Pile

Abstract

The quality of foundation pile is greatly influenced by human factors, and quality assessment is delayed. This paper introduces a new evaluation system based on Internet of Things (IoT) monitoring data of the foundation pile construction process. First, an IoT monitoring system of foundation pile construction process quality is established to monitor the key parameters for quality control in the foundation pile construction process, such as pile length, position, verticality, water–cement ratio, grouting volume, drilling/lifting speed, etc. Next, the absolute gray relational degree analysis method and the analytic hierarchy process (AHP) entropy-weighted combination weighting method are used to divide the monitoring data into different levels and determine the weight coefficients for quality indicators during foundation pile construction. Last, the IoT monitoring and evaluation system of the foundation piles construction process quality is applied to engineering. The results indicate that the monitoring system is convenient and efficient, and the quality evaluation method is reliable. The construction process quality of cement-mixing piles is rated as excellent. The construction process quality of bored piles Z0103 and Z0232 is excellent, and pile Z0012 is qualified.

1. Introduction

Foundation pile is a foundation form with high bearing capacity, a wide application range, and various types. It is widely used in high-rise buildings, ports, bridges, and other projects. Foundation pile construction has many steps, and quality control is difficult.

Bored piles and cement–soil mixing piles are common types of foundation piles [1,2]. Cement–soil mixing pile is commonly used to improve the soft soil foundation as it is simple, highly effective, and low cost [3,4,5,6,7]. As a foundation form, the bored pile is broadly exploited in highways, bridges, sluices, and other engineering fields due to its strong adaptability, moderate cost, and simplicity [8,9,10,11]. However, the construction quality of bored piles is often difficult to control due to the high concealment of the construction site and the low level of automation of the construction equipment. At present, the construction of piles in China is still mostly completed by manual operation. Thus, the construction quality depends on the experience and responsibility of the operators and is often difficult to control. As a result of the uneven quality of supervisors and inadequate staffing, the outcome of on-site construction supervision is often unsatisfactory. The sampling test after construction is based on the idea of probability theory, that is, the quality of several piles is used to judge the construction effect of all piles on the whole site. However, due to the dark box operation, the selection of the test piles is non-random, and the test results cannot reflect the entire construction quality of the site. Since the quality inspection happens after the construction, remediation, or rework is required if the construction is unqualified, which will affect the construction progress. Additionally, the currently used post-construction sampling inspection methods are mainly laboratory-based and on-site tests such as load tests, coring tests, and unconfined compressive strength tests of core samples, which are time consuming and labor intensive [12,13,14,15,16,17,18,19,20].

The IoT technology has become a key technology to solve the interaction and integration between physical entities and data virtualization in intelligent manufacturing [21,22]. Through the integration of artificial intelligence, IoT, and other technologies, it is applied in equipment monitoring, pile operation and maintenance, and dispatch operation. The integration of new IT technologies, such as big data, meets the needs of intelligent manufacturing and cyber–physical systems. The IoT pile model is established based on the 5D model of the IoT [23,24,25]. Through the real-time interaction of data and operations between the physical pile and the IoT pile, the data integration and business integration in the whole lifecycle of the pile are improved, and the process of design, construction, operation, and maintenance is realized in the pile. By installing various sensors, one can monitor the cement mixing pile and related equipment in real time. Surrounding security, through the SCADA system and electronic equipment instruments to transmit voltage and other signals while receiving the process analyzer, one can achieve real-time monitoring of the construction status of cement mixing piles. On this basis, the real-time data of the pile is uploaded to the IoT pile and pile service system, the data of the pile is converted, cleaned, and packaged, and the data of the pile body is integrated [26,27].

Quality evaluation of the construction process is closely linked to quality inspection. With the continuous improvement of testing methods, construction data that were previously unobtainable can now be extracted, providing a basis for better evaluation of the pile foundation construction process. He J. combined fuzzy synthesis with AHP to study a bridge foundation pile construction quality evaluation method based on acoustic wave reflection pile foundation monitoring results. Peng J. and others proposed a pile foundation quality evaluation method based on the BP neural network artificial intelligence algorithm. Xu H. et al. proposed a core-drilling pile quality evaluation method based on AHP and gray theory. Qin Y. et al. used the entropy weight method to evaluate the bearing capacity of PHC pipe piles, concrete durability, and other indicators. Momeni et al. [28,29,30] developed a hybrid model by combining artificial neural networks (ANN) with a genetic algorithm (GA) for predicting the bearing capacity of piles. Similarly, other AI techniques have also been successfully implemented for predicting the ultimate bearing capacity of different types of piles using ANN: support vector machines (SVM), relevance vector machines (RVM), multivariate adaptive regression splines (MARS), extreme learning machines (ELM) and functional networks (FN) [31,32,33,34]. Apart from the load-bearing capacity of piles, ANN has also been used for predicting the settlement of piles based on cone penetration tests (CPT) [35,36,37,38,39].

This paper takes cement–soil mixing piles and bored piles as the research objects. On the basis of the IoT monitoring system and management platforms for the construction process of foundation piles, the evaluation system of pile foundation construction process quality based on the multi-objective fuzzy synthesis and AHP is developed and applied to engineering.

2. IoT Monitoring Platforms for the Pile Construction Process

2.1. Monitoring System and Information Platform

(1) Monitoring System

Figure 1 shows overal structure diagram of the monitoring system The quality monitoring of foundation pile is developed to transmit, store, and browse data and evaluate quality by Internet and IoT technologies. The sensor’s data are connected to a monitoring host, and the communication module of the monitoring host is responsible for uploading information to the cloud server, which can be accessed via client-side and mobile-end interfaces.

The IoT monitoring system is based on a combined wired and wireless transmission format. Instruments such as rangefinders, hydrometers, and dual-axis angular displacement sensors are connected via RVV cable-type RS485 bus to form a group of address-independent slave devices. These devices are networked with the monitoring host using the MODbus protocol and establish communication connections. Once the network connection is successfully established, the monitoring host sends read/write data requests to the various sensors.

Upon receiving these requests, the sensors send corresponding index data to the main station. After reading the data, the monitoring host displays it and transmits the data externally via an antenna and a SIM card. The antenna converts electrical signals into radio frequency signals and transmits them to the base station via radio waves. The SIM card sends data signals to the base station using protocols such as GSM, UMTS, and LTE. Upon receiving signals from the antenna or SIM card, the base station converts them into digital signals and forwards them to the core network for processing. The core network processes the received signals, including decoding, decrypting, and routing, before forwarding the signals to the target device, which is the server.

Client terminals, such as PCs or mobile devices, establish TCP connections with the server via the TCP protocol and send data reading and upload requests to the server using the TCP protocol. After receiving the request data, the server processes it and returns the corresponding data as needed. During data transmission, the TCP/IP protocol performs operations such as packet segmentation, ordering, and retransmission to ensure data integrity and reliability.

(2) Information platform

The information platform included four modules: project management, data management, alarm management, and quality evaluation, which include features such as inputting design data and evaluation standards, real-time monitoring of construction data, quality analysis during the pile-driving process, and generation of inspection reports, as shown in Figure 2. The information platform includes a client and a mobile terminal.

The client application included seven main functions are project mapping, equipment management, project management, alarm management, data management, quality evaluation, and personnel management. Design drawings, construction specifications, and personal information for engineering projects can be added through the management backend. The quality evaluation module evaluates completed pile construction processes by scoring them based on the actual parameters monitored by the IoT monitoring system, comparing them to the specifications for pile foundation construction. Each pile location is assigned a score ranging from 0 to 100, corresponding to evaluations of excellent, qualified, or unqualified. The alarm management module allows for the setting of construction requirements, referred to as alarm thresholds, based on pile foundation construction specifications, on-site construction conditions, and project circumstances. When the actual values of monitored indicators during construction exceed these thresholds, alarms are triggered. The personnel management module is used for assigning roles to project personnel and managing on-site construction staff. Different system permissions are allocated to workers with different responsibilities, ensuring the security of engineering data and the orderly operation of project management structures.

The mobile application offers four main functions: map monitoring, data management, alarm management, and project management, as shown in Figure 3. Construction management personnel can use this platform to access pile construction information at any time.

2.2. IoT Monitoring System for Cement–Soil Mixing Pile Construction

2.2.1. Quality Control Parameters for Cement–Soil Mixing Pile Construction

Construction quality parameters are crucial for controlling the quality of cement–soil mixing pile construction. Key factors affecting the quality of cement–soil mixing pile construction include the uniformity of mixing during construction and the amount of cement admixture. Mixing uniformity is mainly influenced by the drilling/lifting speed and the rotation speed of the mixing head. The amount of cement admixture is primarily affected by the grouting volume and the water–cement ratio, with the grouting volume determined by the drilling/lifting speed and grouting speed, and the water–cement ratio determined by the slurry density and viscosity.

The main construction processes of cement–soil mixing piles include machine positioning, pre-mixing and submerging, pre-mixing and lifting, re-mixing and submerging, and re-lifting, among others. The construction processes and quality control indicators are shown in Figure 4. Based on the construction technology of cement–soil mixing piles and the conditions at the construction site, this paper combines the assurance items, general items, and allowable deviation items in the quality inspection standards for cement–soil mixing piles. It outlines monitoring parameters related to pile location, pile length, pile position, verticality, grouting rate, drilling/lifting speed, mixing head speed, slurry density, slurry viscosity, and others. These parameters are comprehensively considered to ensure the quality of the cement–soil mixing pile construction process. The control standards and alarm values for these indicators are provided in

Table 1. Quality inspection standards for cement–soil mixing piles.

Serial Number	Key Parameter	Standard Value	Alarm Value
1	Pile deviation	50 mm	>50 mm
2	Spray volume	Design values ±1%	≠Design values ±1%
3	Water–cement ratio	Design values ±0.05 g/cm3	≠Design values ±0.05 g/cm3
4	Trip and lift speed	Design values ±0.05 m/min	≠Design values ±0.05 m/min
5	Borehole depth	Not less than the design value	<Design values
6	Verticality	1%	>1%

.

2.2.2. Sensors for Cement–Soil Mixing Pile Construction

The monitoring sensors fixed on the drilling system contained a position sensor, a verticality sensor, and a depth sensor for monitoring the retracting velocity and verticality of the drill, and the position of slurry spraying. The monitoring sensors fixed on the feeding system contained an electromagnetic flowmeter for monitoring the flow rate of grouting. An inserted sound fork was used in the pulping system for real-time monitoring of the water–cement ratio.

Table 2. Sensors for cement–soil mixing piles.

Serial Number	Key Parameter	Method	Sensor
1	Pile deviation	GNSS	GNSS receiver
2	Spray volume	Monitoring slurry flow	Electromagnetic flowmeter
3	Water–cement ratio	Mud-induced vibration	Plug in tuning fork
4	Trip and lift speed	Depth meter velocity measurement	Depth sensor
5	Borehole depth	Depth gauge	Depth sensor
6	Verticality	Inclination sensor	Dual-axis angle sensor

shows the information about the sensors. Figure 5 shows the installation of monitoring sensors.

2.3. IoT Monitoring System for the Construction of Bored Pile

2.3.1. Quality Control Parameters for Forward/Reverse Circulation Bored Pile Construction

Construction process quality parameters are crucial for controlling the quality of bored pile construction. This paper focuses on the Internet of Things (IoT) monitoring and evaluation of the quality of upward/downward circulation drilling cast in situ piles. The main construction processes for upward/downward circulation bored pile construction include stakeout of pile locations, site leveling, sinking, installation of casings, erection of pile frames, positioning of drilling machines, slurry preparation, drilling, fabrication and installation of steel reinforcement cages, hole cleaning, underwater concrete pouring, and casing extraction, among others. The construction processes and quality control indicators for bored pile construction are shown in Figure 6. The chosen monitoring parameters include verticality, pile length, pile location deviation, as well as slurry-specific gravity and viscosity. The control standards and alarm values for these parameters are presented in

Table 3. Quality inspection standards for bored piles.

Serial Number	Index	Standard Value	Alarm Value
1	Pile deviation	100 + 0.01 H	>100 + 0.01 H
2	Borehole depth	100% Design value	<80% Design value
3	Verticality	1%	≥1%
4	Mud weight	1.15–1.20	<1.15 or >1.20
5	Mud viscosity	16~21 s	<16 s or >21 s

.

2.3.2. Sensors for the Construction of Bored Pile

The drilling construction of forward/reverse circulation bored piles adopts the drilling construction of continuous fixed-length drill pipes. The laser range finder is welded and installed on the top of the drill pipe. The real-time distance of the bearing horizontal disk estimates the drilling length of a single drill pipe. The infrared counter is installed at the horizontal position of the slewing bearing horizontal plate to record the number of consecutive drill pipes. The real-time measurement data can also estimate the drilling rate of the drilled piles.

The monitoring sensors fixed on the drilling system contained a position sensor and a verticality sensor for monitoring the verticality of the drill and the position of slurry spraying. An inserted sound fork was used in the feed system for real-time monitoring of the slurry density.

Table 4. Sensors for the bored piles.

Serial Number	Key Parameter	Method	Sensor
1	Pile deviation	GNSS	GNSS receiver
2	Borehole depth	Pile extension times	Electromagnetic flowmeter
3	Verticality	Inclination sensor	Dual-axis angle sensor
4	Mud weight	Mud-induced vibration	Plug in tuning fork
5	Mud viscosity	Rotating torque principle	Online rotary viscometer

shows the information about the sensors. Figure 7 shows the installation of monitoring sensors.

3. Evaluation of Foundation Pile

A quality evaluation indicator system is established based on IoT monitoring data. The absolute gray correlation analysis method is employed to determine the evaluation standard thresholds for construction quality indicators. The AHP, combined with entropy weighting, is used to assign weights to different construction quality indicators. Finally, the fuzzy comprehensive evaluation method is applied to conduct construction quality evaluation based on IoT monitoring data. This paper requires experts to be engaged in the field of underground engineering construction for at least 10 years.

3.1. Evaluation Indicators

The quality evaluation indicators for pile construction include pile position deviation, drilling depth, verticality, mud-specific gravity, and mud viscosity. According to different quality control objects, it is divided into three categories: pile structure B1, pile position deviation B2, and mud index B3, where B1 = (drilling speed, drilling depth, verticality), B2 = (pile position deviation), B3 = (sand content, mud viscosity, mud specific gravity, mud flow rate).

3.2. Evaluation Thresholds

To handle the fluctuations in construction quality monitoring data, the gray theory is applied. This involves analyzing how closely the quality monitoring values correlate with the specified thresholds, thereby determining how accurately the quality data reflect the actual conditions [40]. The calculation process is as follows:

(1) Determining data sequences

Collect real-time quality monitoring data at minute intervals. Suppose there are n data points for quality indicator i. These data points together constitute the quality monitoring data sequence Xi.

$$X_i=x_i\left(1\right),x_i\left(2\right),\cdots ,x_i\left(n\right)$$

(1)

For each quality monitoring indicator, there exists a corresponding set of specified values denoted as x₀ (n). These specified values collectively form the specified value sequence X₀.

$$X_0=x_0\left(1\right),x_0\left(2\right),\cdots ,x_0\left(n\right)$$

(2)

If the specified values are given as a range, the reference value is taken as the average of the upper and lower limits of that range.

(2) Calculating correlation [41].

This paper departs from traditional models that only consider the area between curves. Instead, it uses methods that calculate both the absolute value of the area and the area difference between the two curves.

Given that x_i and x₀ have the same length, let:

$$s_i-s_0={\displaystyle {\int }_1^n|x_i-x_0|dt}$$

(3)

$${\rho }_{i0}={\displaystyle \frac{1}{1+|s_i-s_0|}}$$

(4)

where ${\rho }_{i0}$ represents how closely the i-th quality indicator monitoring value during foundation pile construction correlates with the specified value. n is the number of monitoring results. i is monitoring times.

3.3. Calculation of Indicator Weight

The weight calculation for indicators is performed using a combination of the AHP and entropy-based weighting. AHP is used to calculate subjective weight values for the indicators, while entropy-based weighting calculates objective weight values for the evaluation indicators. Finally, these weights are combined to determine the comprehensive weight.

(1) Subjective Weight Calculation

The AHP is employed for the subjective weight calculation. This process includes the following steps: constructing the indicator hierarchy, creating pairwise comparison matrices, computing the weight coefficient vectors for these matrices, conducting a consistency check, and ultimately ranking the hierarchy.

(2) Objective Weight Calculation

Entropy-based weighting, on the other hand, is used to calculate objective weights for the evaluation indicators. The process involves the following steps:

① Constructing pairwise comparison matrices for m piles and r indicators:

$$D_{pi}=\left[{\rho }_{pi}\right]p=1,2,\cdots ,m;i=1,2,\cdots ,r$$

(5)

where ${\rho }_{pi}$ is ${\rho }_{i0}$ of the i index of the p pile foundation.

② Normalizing the judgment matrix ${\dot{\rho }}_{pi}$

③ Among the r evaluation indexes, the entropy of the i evaluation index is

$$S_i=-{\left(\mathrm{In} r{\displaystyle \sum _{i=1}^r\left({\dot{\rho }}_{pi}\mathrm{In} {\dot{\rho }}_{pi}\right)}\right)}^{-1}$$

(6)

④ The entropy weight of the i-th evaluation index is

$$w_i^{″}={\displaystyle \frac{1-S_i}{r-{\displaystyle \sum _{i=1}^r{S}_i}}}$$

(7)

where $w_i^{″}$ represents the objective weight value for the i-th evaluation indicator, and the objective weight vector is denoted as W″ = (w″₁, w″₂, …, w″_n)T.

(3) Comprehensive Weight Calculation

To avoid the drawbacks associated with either purely subjective or purely objective weightings, a linear combination of weights is used. Let the comprehensive weight vector for the various indicators be denoted as W = (w₁, w₂, …, w_r)T. Then, the comprehensive weight for the i-th indicator, w_i, is calculated as follows:

$$w_i=(1-\delta ){w^{\prime }}_i+\delta {w^{″}}_i$$

(8)

where $\delta$ is a weight coefficient that ranges between 0 and 1. $\delta$ = 0 indicates that the evaluation relies solely on subjective judgment. $\delta$ = 1 indicates that the evaluation relies solely on objective evaluation. $\delta$ = 0.5 indicates that both subjective and objective evaluations are equally relevant.

If there are differences in the accuracy of subjective and objective evaluations, $\delta$ can be adjusted accordingly to reflect this discrepancy.

3.4. Evaluation of the Construction Process

The construction process quality evaluation is carried out using the fuzzy comprehensive evaluation method, which is based on fuzzy mathematics. It involves constructing membership functions and making a comprehensive judgment through methods such as the maximum membership principle or weighted average. The steps are as follows:

(1)

Determine the Factor Set: Based on the established evaluation indicator system and the evaluation standard thresholds for construction process quality indicators, the relevance of the monitoring values of each evaluation indicator to their respective normative values is represented as: U = (u₁, u₂, …, u_i).

(2)

Determine the Scheme Set: The quality of the construction process indicators is categorized as excellent, qualified, or unqualified.

(3)

Establish Membership Functions: The relationship between the factor set and the scheme set is expressed through membership functions. In this paper, trapezoidal distribution membership functions are selected. The distribution curve of the membership function is shown in Figure 8.

$$A_{v3}\left(x\right)=\left\{\begin{array}{cc}0& x<c\\ {\displaystyle \frac{x-c}{d-c}}& c\le x\le d\\ 1& d<x\end{array}\right.$$

(9)

outstanding

$$A_{v2}\left(x\right)=\left\{\begin{array}{cc}{\displaystyle \frac{x-a}{b-a}}& a\le x\le b\\ 1& b\le x\le c\\ {\displaystyle \frac{d-x}{d-c}}& c\le x\le d\\ 0& x<a \mathrm{or} d<x\end{array}\right.$$

(10)

Up to standard

$$A_{v1}\left(x\right)=\left\{\begin{array}{cc}1& x<a\\ {\displaystyle \frac{b-x}{b-a}}& a\le x\le b\\ 0& b<x\end{array}\right.$$

(11)

Below standard

a, b, c, and d represent the boundary values for the distribution of the membership function, and these boundary values may vary for different indicators and are determined by the standard values. The fuzzy evaluation matrix for r indicators can be represented as:

$$R_i=\left[\begin{array}{ccc}{r}_{11}& r_{12}& r_{13}\\ r_{21}& r_{22}& r_{23}\\ ⋮& ⋮& ⋮\\ r_{r1}& r_{r2}& r_{r3}\end{array}\right]$$

(12)

(4)

Calculate the comprehensive evaluation vector

$$H_p=W\circ R=\left(h_1,h_2,h_3\right)$$

(13)

where $H_p$ is the comprehensive evaluation vector. W is the comprehensive weight matrix. $\left(h_1,h_2,h_3\right)$ are the membership degrees of the construction process quality indicators relative to the first, second, and third evaluation grades. “$\circ$” represents the fuzzy composition operator.

4. Field Test

4.1. Field Test of Cement–Soil Mixing Pile

4.1.1. Test Result of Cement–Soil Mixing Pile

The experiment site is located in Yangzhou, Jiangsu Province, China. The predominant soil composition at the site comprises Quaternary deposits, specifically fine-grained clay, silty loam, and clayey silt, resulting from Quaternary alluvial deposits of the Upper and Middle Yangtze River.

For this project, 600 mm diameter cement mixing piles were utilized. The piles were solidified using 42.5 OPC as the curing agent, with a cement admixture rate of 18%. The water–cement ratio was 0.5. Construction followed the four-mixing and three-injection technique, with the lifting rate strictly controlled not to exceed 1 m/min.

The test piles were selected randomly in various areas of the project site, which were identified as Z0102, Z0130, and Z0062. The layout of the cement–soil mixing test pile is shown in Figure 9.

Throughout the testing process, real-time data were continuously collected by various monitoring systems and subsequently transmitted to a cloud-based platform. The results obtained during the testing phase are presented in Figure 10. Results revealed that the actual values of all parameters fluctuated within the allowable limits specified by the standards, demonstrating that the construction results met the required standards. The IoT monitoring systems operated seamlessly, ensuring stable data transmission. The on-site testing has unequivocally demonstrated that this system is a powerful asset for construction management personnel, providing them with effective control and oversight of the site.

4.1.2. Construction Quality Evaluation of Cement–Soil Mixing Pile

(1) Index system

Figure 11 shows Quality evaluation index system for cement–soil mixing pile construction process.

(2) Index analysis

Table 5. Correlation between measured values and normative values of various indicators of Cement–Soil Mixing Pile.

Pile Position	Degree of Relevance	Borehole Depth	Perpendicularity	Pile Deviation	Water-Cement Ratio	Cement Slurry Flow	Drill Down Speed
Z0102	$s_i-s_0$	0.037	7.289	0.162	0.346	1.432	0.576
	${\rho }_{i0}$	0.964	0.121	0.861	0.743	0.411	0.635
Z0130	$s_i-s_0$	0.350	8.534	0.104	0.418	1.381	0.420
	${\rho }_{i0}$	0.741	0.105	0.906	0.705	0.420	0.704
Z0062	$s_i-s_0$	0.089	7.911	0.097	0.219	1.680	0.331
	${\rho }_{i0}$	0.918	0.112	0.912	0.820	0.373	0.751

shows correlation be tween measured values and normative values of various indicators of Cement–Soil Mixing Pile.

(3) Index weight

Table 6. Combination weights of quality evaluation indicators for the construction process.

	Borehole Depth	Perpendicularity	Pile Deviation	Water-Cement Ratio	Cement Slurry Flow	Drill Down Speed
Subjective weights $W^{\prime }$	0.254	0.153	0.102	0.136	0.136	0.220
Objective weighting $W^{″}$	0.221	0.143	0.143	0.146	0.146	0.200
Combined weights $W$	0.238	0.148	0.122	0.141	0.141	0.210

shows combination weights of quality evaluation indicators for the construction process.

(4) Comprehensive Evaluation Results

The comprehensive evaluation calculations for the construction process quality of different piles yielded the following results:

$$H_{Z0102}=W\circ R=\left(0.531,0.413,0.056\right)$$

$$H_{Z0130}=W\circ R=\left(0.608,0.299,0.093\right)$$

$$H_{Z0062}=W\circ R=\left(0.682,0.300,0.018\right)$$

According to the “take the larger value” principle, the construction process quality evaluation results for the tested foundation piles of cement mixing piles are as follows: piles Z0102, Z0130, and Z0062 all achieved excellent construction process quality. The order of the three piles was Z0062 > Z0130 > Z0102.

4.2. Field Test of Bored Piles

4.2.1. Test Result of Bored Piles

The field test site is located in Nantong, Jiangsu Province, China. The foundation mainly consists of Quaternary alluvial deposits, including sandy silt, silty sand with sandy silt interlayers, silt, sandy silt with sandy interlayers, and silty clay of unified origin. The project used 350 bored piles with a diameter of 600 mm and an effective pile length of about 27 m. The layout of test bored piles is shown in Figure 12.

Two test piles were selected from each of the three areas, resulting in a total of nine single piles with the IDs of Z0012 and Z0103. Figure 13 shows the construction data monitoring curve of test piles. Results revealed that the actual values of various indicators exhibited fluctuations within the permissible limits during the construction process, and the construction results met the required standards. The monitoring system operated normally during construction, ensuring stable data transmission.

4.2.2. Construction Quality Evaluation

(1) Index system

Based on the IoT monitoring data for the construction process quality of foundation piles, the correlation between the actual construction process data and the specified values was calculated using Formulas (3) and (4), as shown in

Table 7. Correlation between measured values and normative values of various indicators of Bored Piles.

Pile Position	Degree of Relevance	Borehole Depth	Verticality	Pile Deviation	Specific Gravity of Mud	Mud Viscosity
Z0012	$s_i-s_0$	0.570	8.225	0.061	1.786	24.843
	${\rho }_{i0}$	0.021	0.108	0.942	0.359	0.039
Z0103	$s_i-s_0$	0.040	8.024	0.145	1.167	37.003
	${\rho }_{i0}$	0.001	0.111	0.873	0.462	0.026
Z0232	$s_i-s_0$	0.470	7.190	0.157	0.958	31.539
	${\rho }_{i0}$	0.017	0.122	0.864	0.511	0.031

.

(2) Index weight calculation

(1) Primary Weights

The project-level judgment matrix is obtained separately for the three project-level criteria, B1 (Pile Body Structure), B2 (Pile Position Deviation), and B3 (Slurry Parameters), and the five indicator-level criteria, C1 (Drilling Depth), C2 (Verticality), C3 (Pile Position Deviation), C4 (Mud Density), and C5 (Mud Viscosity), based on expert ratings. The judgment matrix is as follows:

$$A=\left[\begin{array}{cccc}& B_1& B_2& B_3\\ B_1& 1& 2& 2\\ B_2& 1/2& 1& 1\\ B_3& 1/2& 1& 1\end{array}\right]$$

The judgment matrix is normalized

$$\dot{A}=\left[\begin{array}{ccc}0.50& 0.50& 0.50\\ 0.25& 0.25& 0.25\\ 0.25& 0.25& 0.25\end{array}\right]$$

According to Formulas (7) and (8), W_A = (0.500, 0.250, 0.250)

Find the maximum eigenvector ${\lambda }_{\max }$

$${\lambda }_{\max }={\displaystyle \sum _{i=1}^3\frac{{\left(A{W}_{Bi}^{\prime }\right)}_i}{r{w}_i^{\prime }}}=3.00$$

The consistency test judgment matrix is obtained by calculating the consistency index

$$CI={\displaystyle \frac{\left({\lambda }_{\max }-r\right)}{\left(r-1\right)}}=0$$

$$CR=0$$

A consistency ratio is CR < 0.1 indicates good consistency in the judgment matrix. Therefore, the project-level criteria weights are as follows:

W_A = (W_B1, W_B2, W_B3) = (0.500, 0.250, 0.250)

Similarly, the indicator-level judgment matrices and weights are as shown in

Table 8. Indicator layer judgment matrix and weights of B1.

B1	Borehole Depth C2	Verticality C3	Weight
Borehole depth C2	1.00	2.00	0.67
Verticality C3	0.50	1.00	0.33

,

Table 9. Indicator layer judgment matrix and weights of B2.

B2	Pile Deviation C3	Weight
Borehole depth C2	1.00	1.00

and

Table 10. Indicator layer judgment matrix and weights of B3.

B3	Mud Viscosity C6	Specific Gravity of Mud C7	Weight
Mud viscosity C6	1.00	1.00	0.50
Specific gravity of mud C7	1.00	1.00	0.50

.

Table 11. Subjective weights of quality evaluation indicators for the construction process of bored pile.

Project Layer	Weight	Metrics Layer	Weight	Primary Weights
B1	0.50	depth C1	0.67	0.33
		Verticality C2	0.33	0.17
B2	0.25	Pile position C3	1.00	0.25
B3	0.25	Specific gravity C4	0.50	0.13
		Viscosity C5	0.50	0.13

shows the subjective weight of the quality evaluation index of the construction process of the cast pile.

(2) Objective weight calculation

The initial matrix is obtained according to the above calculation results

$$D=\left[\begin{array}{cccccc}& C_1& C_2& C_3& C_4& C_5\\ Z0012& 0.942& 0.108& 0.359& 0.039& 0.021\\ Z0103& 0.873& 0.111& 0.462& 0.026& 0.001\\ Z0232& 0.864& 0.122& 0.511& 0.031& 0.017\end{array}\right]$$

Normalized judgment matrix

$$\dot{D}=\left[\begin{array}{ccccc}0.35& 0.32& 0.27& 0.40& 0.53\\ 0.33& 0.32& 0.35& 0.27& 0.04\\ 0.32& 0.36& 0.38& 0.32& 0.44\end{array}\right]$$

The objective weights for evaluation indicators of drilled cast-in-place piles are as shown in

Table 12. Objective weights of quality evaluation indicators for the construction process of bored pile.

	Borehole Depth	Verticality	Pile Deviation	Specific Gravity of Mud	Mud Viscosity
si	0.51	0.68	0.68	0.68	0.67
Objective weighting	0.276	0.179	0.179	0.182	0.183

.

(3) Combined Weights

The combined weights are calculated with β = 0.5, as shown in

Table 13. Combination weights of quality evaluation indicators for the construction process of bored pile.

Weight	Borehole Depth	Perpe Ndicularity	Pile Deviation	Specific Gravity of Mud	Mud Viscosity
Subjective weights $W^{\prime }$	0.330	0.170	0.250	0.125	0.125
Objective weighting $W^{″}$	0.276	0.179	0.179	0.182	0.183
Combined weights $W$	0.303	0.175	0.215	0.154	0.154

.

(4) Quality Evaluation of Pile Construction

Process Based on the construction standards for drilled cast-in-place piles and the quality requirements of site management personnel, the limit values for the membership degree distribution functions are determined, as presented in

Table 14. Boundary value of membership degree distribution function.

Bounds Value	Borehole Depth	Perpendicularity	Pile Deviation	Specific Gravity of Mud	Mud Viscosity
d	0.9	0.105	0.48	0.086	0.05
c	0.68	0.085	0.4	0.063	0.03
b	0.49	0.065	0.32	0.041	0
a	0.33	0.045	0.24	0.019	−0.05

.

Calculate the fuzzy synthesis matrix

$$R_1=\left[\begin{array}{cccc}& v_1& v_2& v_3\\ C_1& 1& 0& 0\\ C_2& 1& 0& 0\\ C_3& 0& 1& 0\\ C_4& 0& 0.895& 0.105\\ C_5& 0& 1& 0\end{array}\right]$$

Pile Z0012

$$R_2=\left[\begin{array}{cccc}& v_1& v_2& v_3\\ C_1& 0.879& 0.121& 0\\ C_2& 1& 0& 0\\ C_3& 0.769& 0.231& 0\\ C_4& 0& 0.332& 0.668\\ C_5& 0& 1& 0\end{array}\right]$$

Pile Z0103

$$R_2=\left[\begin{array}{cccc}& v_1& v_2& v_3\\ C_1& 0.838& 0.162& 0\\ C_2& 1& 0& 0\\ C_3& 1& 0& 0\\ C_4& 0& 0.533& 0.467\\ C_5& 0& 1& 0\end{array}\right]$$

Pile Z0232

The comprehensive evaluation results for the construction process quality of different piles are as follows:

$$H_{Z0012}=W\circ R=\left(0.478,0.506,0.016\right)$$

$$H_{Z0103}=W\circ R=\left(0.606,0.292,0.103\right)$$

$$H_{Z0232}=W\circ R=\left(0.643,0.285,0.072\right)$$

Based on the principle of selecting the highest evaluation result, the construction process quality evaluation results for the tested piles of bored piles are as follows: pile Z0012 has a qualified construction process quality, while piles Z0103 and Z0232 have excellent construction process quality. The order of the three piles was Z0232 > Z0103 > Z0012.

5. Economic Analysis

5.1. Direct Economic Benefits

Traditional pile foundation construction relies on manual layout, monitoring, and recording, requiring the collaboration of multiple surveyors. In contrast, the IoT monitoring system enables real-time data collection, reducing the need for on-site layout personnel by over 60%, eliminating manual inspection steps, and shortening project timelines. The system monitors parameters such as grout volume in real time, preventing over-pouring or under-pouring issues. It reduces the concrete over-pour rate from 8% in traditional construction to within 3%, saving approximately 15% in material costs per pile.

5.2. Indirect Economic Benefits

Traditional pile foundation issues, such as tilted holes and insufficient depth, result in a rework rate of 10% to 15%. The IoT monitoring system enhances the first-time pass rate to over 98% through intelligent monitoring of verticality (e.g., deviation ≤ 5 mm per meter), avoiding rework losses. The cloud platform enables real-time sharing of construction data, reducing management response time from hours to minutes and lowering coordination costs. Automated report generation replaces manual statistics, saving 30% in management hours.

Traditional pile foundation inspections require personnel to enter boreholes, posing higher risks. The monitoring device can reduce the accident rate, reduce the number of on-site personnel, and directly avoid the risk of mechanical injury by replacing manual detection. The traditional reliance on empirical judgment can easily lead to insufficient bearing capacity. The IoT monitoring system can avoid the deviation of design parameters by analyzing the monitoring data in real time. Data link certificates (such as grouting volume and verticality records) provide a tamper-proof basis for acceptance and reduce disputes.

5.3. Cost Composition Analysis

The application costs of IoT monitoring technology primarily include initial investments (sensors, control terminals, etc., accounting for 60% to 70% of total investment), such as approximately RMB 50,000 to 100,000 per pile machine monitoring system. Operational costs include data transmission fees and platform maintenance fees (approximately 10% of the initial investment annually). As shown in

Table 15. Economic comparison of two construction methods.

Cost Item	Traditional Construction/RMB	IoT Monitoring Construction/RMB
Labor	350	200
Materials	1000	850
Equipment	650	750
Total Cost	1900	1800

, taking a 10-m long, 800 mm diameter cement–soil mixing pile as an example, the cost comparison between using the IoT monitoring system and traditional construction methods reveals savings of RMB 150 in labor costs, RMB 150 in materials, and an increase of RMB 100 in equipment costs, resulting in a net saving of RMB 100.

The economic benefits of IoT monitoring technology in pile foundation construction are characterized by “short-term investment, long-term returns,” with its value chain encompassing direct cost savings, risk avoidance, and data-driven value creation.

By optimizing resource allocation, reducing waste, and improving quality and efficiency, IoT monitoring technology ultimately achieves cost reduction across the entire project lifecycle. As sensor prices decrease and 5G networks become more widespread, the economic benefits of IoT in pile foundation construction will become even more pronounced.

6. Conclusions

This study focused on the IoT monitoring and evaluation of pile construction, applying it to cement-mixing piles and bored piles. The main conclusions are as follows:

(1)

The key control parameters for cement-mixing pile construction include pile position deviation, grout volume, water–cement ratio, drilling and lifting speed, verticality, and pile length.

(2)

The key control parameters for bored pile construction include verticality, pile length, pile position deviation, mud density, and mud viscosity.

(3)

The structure of the IoT monitoring system for pile construction includes three main components: on-site data acquisition, data transmission, and monitoring modules. Field tests demonstrated that the system operates stably, data collection is reliable, data transmission is stable, and the cloud platform effectively receives, displays, and processes data on time.

(4)

Based on the field test results of cement-mixing piles, the construction process quality of cement-mixing piles is all rated as excellent, and the order is Z0062 > Z0130 > Z0102.

(5)

Based on the field test results of bored piles, pile Z0012 has a qualified construction process quality, while piles Z0103 and Z0232 have excellent construction process quality. The order of the three piles was Z0232 > sZ0103 > Z0012.