[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-128099-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-128099-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","use-of-machine-learning-methods-to-obtain-a-reliable-predictive-model-for-resilient-modulus-of-subgrade-soil-project-summary","Use of Machine Learning Methods to Obtain a Reliable Predictive Model for Resilient Modulus of Subgrade Soil - Project Summary","","This project develops a machine learning–based predictive model to estimate the resilient modulus (MR) of subgrade soils, a key input for pavement design. It addresses the high cost, time burden, and intensive resource requirements of INDOT’s repeated load triaxial testing method (AASHTO T307). Using existing INDOT MR test data and soil index properties, the work performs literature review, data cleaning, exploratory analysis, anomaly detection, and model training/validation, supported by curve fitting to refine constitutive model coefficients.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/use-of-machine-learning-methods-to-obtain-a-reliable-predictive-model-for-resilient-modulus-of-subgrade-soil-project-summary/128099/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/use-of-machine-learning-methods-to-obtain-a-reliable-predictive-model-for-resilient-modulus-of-subgrade-soil-project-summary/128099.png","ImageObject",300,407,{"name":42,"@type":43},"Riley","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-09-25","2026-08-05",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",9,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does the project address in estimating resilient modulus (MR)?","Question",{"text":63,"@type":64},"The project targets the inefficiency of traditional MR estimation, where repeated load triaxial testing is costly, time-consuming, and resource-intensive.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What inputs are used to build the machine learning model?",{"text":68,"@type":64},"The model uses existing INDOT MR test data together with local soil index properties and other laboratory-related measures from compiled samples.",{"name":70,"@type":61,"acceptedAnswer":71},"How were the predictive models evaluated to ensure reliability?",{"text":72,"@type":64},"The workflow includes rigorous testing and validation by training on a portion of the data and validating on a separate set to check generalization to new data.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},128099,1785944787,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":26,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":80,"read_time":97},687207022233,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","| Joint transportation research program\u003Cbr>Principal Investigators: Sara Khoshnevisan, University of Cincinnati, [sara.khoshnevisan@uc.edu](sara.khoshnevisan@uc.edu), 513.556.5456 Mehdi Norouzi, University of Cincinnati, [norouzmi@ucmail.uc.edu](norouzmi@ucmail.uc.edu), 513-934-3435 Program Office: [jtrp@purdue.edu](jtrp@purdue.edu), 765.494.6508, [www.purdue.edu/jtrp](www.purdue.edu/jtrp)[ ](www.purdue.edu/jtrp)[Sponsor:](Sponsor: Indiana Department of Transportation)[ Indiana Department of Transportation](Sponsor: Indiana Department of Transportation), 765.463.1521 |\n| --- |\n| SPR-4714 2024\u003Cbr>Use of Machine Learning Methods to Obtain a Reliable Predictive Model for Resilient Modulus of Subgrade Soil |\n\nIntroduction  \nThis project aimed to develop an advanced predictive model utilizing machine learning algorithms to improve the efficiency of estimating the resilient modulus (MR) of subgrade soils, which is a critical factor in pavement design. Currently, the Indiana Department of Transportation (INDOT) relies on the repeated load triaxial testing method prescribed by AASHTO T307, which, while effective, is resource-intensive, costly, and time-consuming.  \nThe objective of this study was to use existing INDOT resources, including MR test data and local soil index properties, to develop a machine learning-based model that accurately predicts the MR of subgrade soils. The goal was to minimize the need for extensive laboratory testing, thereby conserving resources and enhancing the efficiency of pavement design processes. By leveraging advanced machine learning techniques, the project intended to create a reliable tool that efficiently and cost-effectively predicted subgrade soil behavior under various stress conditions.  \nThe project followed a structured approach to develop predictive models for the resilient modulus MR of subgrade soil that encompassed several key stages. It began with a comprehensive literature review that examined existing research on MR estimation and prediction, focused on the strengths and limitations of various methods, and identified essential features for model development.  \nDue to the diverse and often noisy nature of geotechnical data, rigorous data cleaning was performed to ensure data quality. This process included removing  \noutliers, correcting errors, and ensuring data consistency, which was crucial for the accuracy and reliability of subsequent analysis. Following data cleaning, exploratory data analysis was conducted to investigate the intricacies of the data from a geotechnical perspective, with a focus on geological and environmental dynamics affecting MR . Anomaly detection was then performed to identify points that significantly deviated from the norm. These anomalies were addressed to prevent potential distortions in the predictive model performance.  \nThe development of the models began with simple linear models and progressed to more complex ones as various machine learning algorithms were evaluated to determine the most effective model for predicting MR based on soil properties.  \nIn addition to ML model development, a curve-fitting method from the SciPy Python library was employed torefine and optimize the coefficients of the constitutive model employed by the repeated load triaxial (RLT) testing equipment at INDOT.  \nFinally, the models were rigorously tested and validated to ensure their effectiveness. This included training on a portion of the data and validating on a separate set to test the model’s ability to generalize to new data. These steps provided a structured approach to handling and analyzing data, developing robust models, and ensuring that the predictions were reliable and valid for practical applications.  \nFindings   \n1. This project led to the development of multiple machine learning models that predicted the resilient  \nmodulus (MR) of subgrade soil. These models used existing INDOT data to estimate MR more efficiently than traditional methods. They were evaluated","cbCaivTHtG22waFL","https://ap.wps.com/l/cbCaivTHtG22waFL","pdf",340755,"English","# Introduction\n# Findings\n## Developed ML predictive models\n## Key soil properties and testing guidance\n## Dataset and future recommendations\n# Implementation","[{\"question\":\"What problem does the project address in estimating resilient modulus (MR)?\",\"answer\":\"The project targets the inefficiency of traditional MR estimation, where repeated load triaxial testing is costly, time-consuming, and resource-intensive.\"},{\"question\":\"What inputs are used to build the machine learning model?\",\"answer\":\"The model uses existing INDOT MR test data together with local soil index properties and other laboratory-related measures from compiled samples.\"},{\"question\":\"How were the predictive models evaluated to ensure reliability?\",\"answer\":\"The workflow includes rigorous testing and validation by training on a portion of the data and validating on a separate set to check generalization to new data.\"}]","Use of Machine Learning Methods to Obtain a Reliable Predictive Model for Resilient Modulus of Subgrade Soil - Project Summary | PDF"]