Mobile application development for estimation of permissible load on shallow and deep foundation using SPT data

The present study demonstrates the development of an Android Application that aims to calculate the allowable bearing pressure for shallow foundations and safe load on pile foundations using the SPT data. The application was built using Android Studio 2020, utilizing XML for the User Interface and Java for the coding. The application offers support for various foundation types, including strip, square, rectangle, and circular shapes for shallow foundations and circular shape for pile foundations. The in-situ SPT data entered by the user was corrected and then processed to calculate soil properties. Subsequently, the bearing pressure for shallow foundation and safe load on the pile was computed adhering to relevant codes. The developed application was verified by comparing the results with already solved examples in the literature. The developed application may be considered under Intelligence in Geotechnics. The created application will be helpful for field engineers to estimate soil parameters and allowable bearing pressure on-site quickly. As a result, it decreases the amount of time and effort necessary for design and thus eliminates the need to refer to tables, codes, and consultants.


Introduction
Before commencing a new construction project, it is essential to thoroughly investigate the soil conditions below the ground to design cost-effective substructures.This site study aims to gather crucial information such as the appropriate foundation type, the position of the water table, laboratory test results for predicting safe load-bearing capacity and settlement, as well as potential soil challenges and environmental issues, if any.Conducting tests on soil samples collected from the site and estimating the bearing capacity can be a time-consuming and labor-intensive process.
Alternatively, a field test can be employed to determine the soil properties.Among the widely used tests for site investigation, the Standard Penetration Test holds significant popularity worldwide.This test involves driving a 50 mm split spoon sampler into the ground at the base of a borehole with a diameter ranging from 100 to 150 mm.The sampler is driven using a 63.5 kg hammer dropped freely from a height of 760 mm.The number of blows required to penetrate each successive depth of 150 mm is recorded.The blows needed for the initial 150 mm penetration are disregarded due to seating error, and the sum of blows required to penetrate from 150 to 450 mm is represented as an N value.The test is conducted at every change in the stratum or at intervals of 1.5 m, whichever is lesser, and the information is recorded in field logs.The N value obtained from the field tests is influenced by various factors, including the type of drilling rigs used and the overburden pressure on the soil [1].The recorded N values need to be corrected for hammer efficiency, drill rod length, sampler lining, borehole diameter, overburden effect, and dilatancy effect [2].After applying various corrections, an average N value for a given stratum and type of footing is computed.The shear strength parameters are correlated using the corrected average N value of the stratum and subsequently, the collapse load of shallow and deep foundations is determined using any of the theories reported in the literature.In this regard, various approaches have been developed in the past [3][4][5][6][7][8] to estimate the bearing capacity of shallow footing resting on the soil.The allowable bearing pressure is calculated using these theories by satisfying shear failure and settlement criteria.Alternatively, Teng's approach [9] can be used to directly obtain the allowable load by considering the average corrected N value of the soil layer.In the case of deep foundations, the ultimate load capacity is determined by considering both the skin friction resistance and point load.Accordingly, the average N value along the pile shaft and base is obtained to determine the shear strength parameters.Berezantsev's approach [10] is followed to determine the bearing load of the pile, while Brom's recommendations [11] are used to estimate skin friction load.
Nowadays, various researcher has started using machine and deep learning techniques in Civil engineering as well for forecasting the compressive strength of the strengthened recycled aggregate concrete [12], compressive strength of brick [13], for spatiotemporal air quality forecasting and health risk assessment [14], to predict the bearing capacity of soil [15][16][17][18][19][20][21][22], air quality [23], tunnelling [24,25] etc. Padmini et al. [15] demonstrated that the Neuro-Fuzzy model outperformed both the Fuzzy and ANN models in predicting the bearing capacity of shallow foundations on cohesionless soils.However, Kalinli et al. [16] found that the performance of the ANN model was better than an Adaptive Neuro-Fuzzy Inference System in determining the ultimate bearing capacity of shallow foundations.Shahnazari and Tutunchian [17] utilized Multigene Genetic Programming (GP) to accurately predict the ultimate bearing capacity of shallow foundations on cohesionless soils.Tsai et al. [18] proposed a genetic programming system (GPS) consisting of weighted genetic programming (WGP) and soft-computing polynomials (SCP), which yielded acceptable prediction accuracy for the ultimate bearing capacity of shallow foundations.Moayedi and Hayati [19] compared various non-linear machine learning and soft computing-based models and found that the feedforward neural network (FFNN) exhibited good agreement with the Finite Element Method (FEM) data.Pakdel et al. [20] investigated the bearing capacity of shallow foundations on anisotropic soil and provided correction factors using limit equilibrium and soft computing techniques.Kashani et al. [21] explored differential algorithm (DE), evolution strategy (ES), and biogeography-based optimization algorithm (BBO) for foundation design optimization, but no algorithm was identified as the most efficient solver.Ahmad et al. [22] developed a GPR model for predicting the ultimate bearing capacity of shallow foundations on cohesionless soil.
With the increasing popularity of smart devices and the ongoing digitalization trend, there is a growing dependency on these devices.However, there is a lack of applications catering specifically to civil engineers, particularly in the field of geotechnical engineering.Researchers have developed certain Android applications for soil classification and slope stability analysis to address this gap.Kumar et al. [26] created an Android application that utilizes the ASTM system to classify soil.The classification process is based on experimentally determined input parameters and commonly used graphs, tables, and flow charts found in the literature.This mobile application accurately categorizes a large number of soils and reduces the occurrence of human errors in soil classification.Similarly, Dutta et al. [27] developed an Android application using Java programming and Android Studio to assess the stability of rock slopes.The application estimates the factor of safety (FOS) assuming plane wedge failure for slope stability analysis.It considers various parameters such as shear strength, unit weight of the rock, geometric characteristics of the slope and fault plane, a surcharge on the slope, seismic acceleration coefficient specific to the location, and the stabilizing force resulting from the presence of rock bolts, if any.
The studies conducted so far have shown a lack of significant progress in developing an Android application specifically designed to estimate the bearing capacity of shallow and deep foundations using Standard Penetration Test (SPT) values.Accordingly, the researchers have developed an application in their current study to address this need.
In this study, an Android application was created to calculate the allowable bearing pressure for shallow foundations and the safe load on pile foundations using Standard Penetration Test (SPT) data.The application considers different plan shapes for shallow foundations, including strip, square, rectangular, and circular footings.For pile foundations, only a circular plan shape was considered.The allowable capacity for both shallow and deep footings was determined based on the relevant provisions of the Indian Standard Code, wherever applicable.A rigorous validation process was conducted to ensure the accuracy and reliability of the application, using selected examples sourced from textbooks and reports.

Objective of work
The goal of this project is to create an Android application that performs the following tasks: 1. Gathering SPT (Standard Penetration Test) data during a test and making necessary corrections to obtain a corrected N value for each soil layer and an average N value for the entire soil stratum up to a significant depth.2. Utilizing the N value to calculate soil properties such as unit weight (γ), relative density (Dr), friction angle (φ), and cohesion (c u ).
3. Taking input from the user regarding the shape, size, and depth of a shallow footing and calculating the allowable bearing pressure for the foundation.4. Collecting information on the length and diameter of a pile foundation and calculating the allowable load capacity of the pile.
Fig. 1 Flow chart to take input for shallow foundation 5. Displaying the corrected N values, relative density, bulk unit weight, friction angle, and cohesion for different soil layers in a visual format.6. Displaying the allowable bearing pressure for a shallow foundation and the allowable load capacity for a pile foundation.

Steps for application development
Before the application was developed, each stage of the process was represented by a flow chart.The flow diagram for shallow foundation and pile foundation is described below.

Shallow foundation
It involves the user selecting the shape of the foundation (strip, square, circular, or rectangular).The process was divided into two steps.In the first step, the user provided input data such as footing type, width, foundation depth, soil type, number of blows during each 150 mm penetration, and groundwater elevation.The recorded Standard Penetration Test (SPT) N value was then corrected based on factors like diameter, sampler type, rod length, hammer efficiency, overburden, and dilatancy [28].The corrected N value was then utilized to calculate soil parameters such as unit weight, friction angle, and cohesion for each soil layer.The flow chart for this first step is illustrated in Fig. 1, and additional details can be found in Bowles [2].In this figure, the term number of layers implies the depths at which the SPT test is conducted.More details regarding the same are provided in the example problem.The bearing capacity was determined for identified soil layer based on the user-provided footing width and depth in the second step of the Shallow Foundation stage.In this regard, the bearing capacity factors (N c , N q , N γ ), including shape factors (s c , s q , s γ ), depth factors (d c , d q , d γ ), and water table correction factors (W' , W"), corresponding to N γ and N q components, were calculated following IS 6403-198 [8].The soil was then checked for local shear failure, and mobilized cohesion and friction angle were determined if necessary.For cohesionless soil, bearing capacity factors (N q , N γ ) were computed using the friction angle (φ), and the safe bearing capacity was determined.The safe bearing capacity was calculated using Teng's [9] approach by Fig. 3 Flow chart to take input for pile foundation considering the corrected average N value.Additionally, the bearing pressure corresponding to a specified settlement was conservatively determined using Teng's [9] approach without considering the depth factor.This approach is expected to yield a similar result to that obtained using IS 8009, Part I (1976) [29].In the case of cohesive soil, bearing capacity factors (N c , N q ) were computed using friction angle (φ) = 0.The flow chart for the second step can be depicted as shown in Fig. 2.

Pile foundation
The flow diagram follows a specific process when the user selects a circular pile foundation in the main activity.The process is divided into three to four parts.In the first step, the user provides input such as pile length, diameter, soil type, number of blows for each 150 mm penetration, and groundwater elevation.Additional data can be added during the test if needed.The SPT N value is then corrected for various parameters, except for overburden and dilatancy effects.Using the corrected N value, soil parameters like unit weight, friction angle, and cohesion are calculated for each soil layer based on the soil type.Overburden and dilatancy corrections are applied to the SPT N value, and the data for each layer is saved for later calculations.The flow diagram for inputting data for the pile Fig. 4 Flow chart to calculate ultimate skin friction for pile foundation foundation is displayed in Fig. 3.In the second step, based on the user-provided pile length, the soil layer at which the bearing capacity needs to be calculated is determined.The unit skin friction resistance is then computed for each layer, considering whether the soil is cohesive or cohesionless.Finally, the ultimate skin friction resistance is calculated by multiplying the unit skin friction resistance with the surface area of the pile in contact with the soil.The flow diagram for calculating ultimate skin friction for pile foundation is shown in Fig. 4. In the third step, the ultimate point load is determined at the location of the pile tip.This step can be visualized in the flow diagram shown in Fig. 5. Finally, the total pile capacity at failure is obtained by adding the ultimate skin friction resistance and point load.The allowable pile load is then calculated using a safety factor of 2.5.

Creation of user interface using XML
The application's main screen serves as the home page and provides options for the user to select the type of footing they want to analyze.The user is directed to the appropriate activity depending on the chosen footing type.In the subsequent activity, the user is required to input the test data from the Standard Penetration Test (SPT) and provide properties of the footing, such as depth, width or diameter, length, etc.The input data must be numerical, and if any data is invalid or left blank, a pop-up message will prompt the user to enter the data correctly.However, users can enter zero if they don't want to input specific data.Finally, the results of the calculations are displayed to the user in another activity.The XML code for inputting layer data, footing properties, and displaying results is depicted in Figs. 6, 7, and 8, respectively.

Application development
The mobile application is built using the Java programming language and Android Studio.It utilizes the flow diagram discussed earlier and integrates it with the XML-based activities created previously.The application's main screen, depicted in Fig. 9, prompts the Fig. 5 Flow chart to calculate ultimate point load for pile foundation user to select the type of footing.Upon choosing the footing type, a second window appears where the user is required to input the width and depth of the foundation, as shown in Fig. 10.The user can tap the ' + ' button to input data for subsequent tests.Once all the Standard Penetration Test (SPT) data is entered, the user needs to input the groundwater table depth and select the 'Solve' option.

Results and discussions
The application was tested and verified using an example focusing on shallow foundations.The example involved various input parameters, including the type of footing (rectangular), dimensions of the footing (5 m × 3 m), and the depth of the groundwater table (3 m). Figure 11 shows the soil profile for the shallow foundation.The problem is solved in the application as follows: The footing type is selected as 'Rectangular Footing' (Fig. 12).Then the footing length, width, and depth of footing are entered as 5 m, 3 m, and 1.5 m, respectively.The SPT test details are also entered one by one.And finally, the depth of the groundwater table is set as 3 m (Fig. 13).After entering all the necessary data, the user pressed the 'solve' button, and the application provided the results, as shown in Fig. 14.This example was also manually solved, and the answer is provided below [30].The calculated N value for a shallow foundation is shown in Table 1.
The net ultimate bearing capacity (q nu ), as per IS 6403-1981 [29], is calculated as per the following formula:
Additionally, the net safe settlement pressure (q sp ) for a settlement of 40 mm, as per Teng [9], is calculated using the formula The allowable bearing pressure will be the least of the net safe bearing capacity (q ns ) and net safe settlement pressure (q sp ), which is 209.63 kPa.This value closely matches the allowable bearing pressure obtained from an application, which is 208.1 kPa, as shown in Fig. 14.Both approaches yield similar results, as shown in Table 2.
Note that according to the flow chart shown in Fig. 1 and data shown in Table 1, this example problem consists of 8 layers.Furthermore, the variation of soil properties with depth (8 layers) is provided in Fig. 14.
The accuracy of the application was also verified using real-field data from an SPT test conducted in Jalandhar, India.The test outcome was used to determine the q sp = 209.63kPa bearing capacity based on shear failure considerations.
The corrected N values for a specific borehole are presented in Table 3.The average N value obtained was 12.60.The assumed depth and width of the square footing were taken as 1.5 m and 2.5 m, respectively.To calculate the net ultimate bearing capacity according to IS 6403-1981 [8], various parameters were considered and depicted in Table 4.The equation used to calculate the net ultimate bearing capacity (q nu ) is as follows: The net safe bearing capacity ( q ns ) = 34.972.5 = 13.99t/m 2 The net safe bearing capacity obtained was 14.34 t/m 2 for the same test data with a mobile application.
It is widely recognized that for cohesionless soils, the bearing pressure is primarily influenced by settlement criteria, whereas for cohesive soils, it is determined by the shear failure criterion.Consequently, when considering sand, the bearing pressure is influenced by factors such as the SPT N value, footing width, and acceptable settlement.Conversely, in the case of clayey soil, the bearing pressure depends solely on the undrained shear strength and the plan shape and embedment depth of the footing.
The developed application is validated for pile foundations using an example.
In this example, the inputs were (i) Pile type = Circular, (ii) Pile diameter = 400 mm, and (iii) Depth of groundwater table = 3 m.The soil profile for this example problem is shown in Fig. 15.This example was also solved manually, and the solution is given below.Table 5 shows the calculation of the N value for the pile foundation.
The following steps were performed in the manual solution to calculate the skin friction for loose sand in pile foundation: 1. Determine the SPT N-values at each depth: The SPT test details were entered in the application, and the N-values were obtained from Table 5. 2. Calculate the effective stress (σ v ′) at each depth: Effective stress is the difference between the total stress and the pore water pressure.In this example, the effective stress was determined appropriately considering groundwater at 3 m depth.3. To determine the skin friction at different depths within a soil layer, Brom's recommendations [11] were followed.The skin friction values were estimated using specific formulas based on the soil type.For cohesionless soil, the skin friction stress (f s ) was calculated as f s = K σ v ′ avg tanδ, where K represents the lateral earth pressure coefficient, σ v ′ avg denotes the average effective stress at the center of the soil layer, and δ represents the friction angle between the soil and the pile interface.
On the other hand, for cohesive soil, the skin friction (f s ) was defined as fs = α c u , where α represents the adhesion factor and cu denotes the undrained shear strength of the soil.These formulas were used to calculate the respective skin friction stress values at different depths within the soil layer.As per Tomlinson's recommendation, the skin friction was limited to a maximum value of 100 kPa.The skin friction capacity for a given layer was calculated by multiplying skin friction stress with the surface area of a layer.4. Sum up the skin friction values at all depths: The individual skin friction values calculated at each depth are summed up to obtain the total skin friction for the pile foundation.The bearing pressure at the pile tip was calculated as q pu = σ v ′ b N q wherein the N q was taken as per Berezantsev's approach [10], where σ v ′ b effective vertical stress at the pile tip considering the arching effect.The estimated bearing stress was limited to 11,000 kPa.The bearing stress was multiplied by the bearing area to arrive at the bearing load.
These steps for manual calculations are described below.

Calculation of skin friction
For Loose Sand     The application then solves the problem by selecting the footing type as 'Circular Pile Foundation' (Fig. 16).Then the pile length and footing width are entered as 14 m and 0.4 m, respectively.The SPT test details are entered one by one.And finally, the depth of the groundwater table is set at 3 m (Fig. 17).The 'Solve' button is pressed when all the information is entered.The obtained result is shown in Fig. 18.
The manually calculated allowable pile load capacity for this example was 864 kN.Whereas the value of    allowable pile load capacity obtained from the application was 870.06 kN, as shown in Fig. 18.The value of allowable pile load capacity obtained from both approaches is almost similar, as shown in Table 6.

Conclusions
This research aimed at creating a user-friendly mobile application using Java programming and Android Studio to estimate the safe bearing capacity of shallow foundations and the allowable load on pile foundations.The application was developed with the intention of providing a convenient tool for field engineers to quickly estimate soil properties and allowable bearing pressure on-site without relying on extensive laboratory tests.The results obtained from the application generally align with manual calculations, although some variations between the two methods may exist.Overall, this project represents an earnest endeavour to assist engineers in reducing the time and effort required for design, as it eliminates the need to consult tables, codes, and experts.

Limitations and future scope
It is worth mentioning that the developed application lacks the functionality for users to input the unit weight and shear strength parameters obtained from laboratory tests.Instead, empirical correlations were utilized to estimate these values.Moreover, the application automatically switches between general and local shear failure modes based on intuitive decision-making rather than relying on user input.Also, the acceptable settlement for sand was fixed at 40 mm.Hence a linear interpolation is required to estimate bearing pressure at any other settlement.Additionally, the safe capacity assessment for piles is solely based on the shear failure criterion, with no provision for evaluating settlement.These aspects can be considered limitations of the present study, suggesting potential areas for future research.

Fig. 2
Fig. 2 Flow chart to calculate allowable bearing pressure for shallow foundation

Fig. 7
Fig. 7 XML file to take input of footing properties

Fig. 8
Fig. 8 XML file to display results

Fig. 9
Fig. 9 Home screen of the application

Fig. 10
Fig. 10 Screen to input data

Fig. 15
Fig. 15 Soil profile for pile foundation

Fig. 17
Fig. 17 Inputting data of pile foundation

Fig. 18
Fig. 18 Results for pile foundation

Table 1
Calculation of N value for shallow foundation

Table 2
Results comparison for rectangular footing

Table 3
Corrected N values for a bore hole

Table 4
Parameter required to determine bearing capacity

Table 5
Calculation of N value for pile foundation Fig. 16 Selecting 'Circular Pile Foundation'

Table 6
Results comparison for pile foundation