Land valuation is a crucial process that involves the comprehensive assessment of specific land parcels to estimate their worth based on market value or predefined value base on valuation standards and regulations. Traditionally, this process has relied on expert involvement, often leading to different value figure for the same land by different valuation officers. While minor variations are considered acceptable, concerns have been raised regarding the lack of transparency, accuracy, and efficiency in the traditional approach. In response to these challenges, this study proposes an innovative system leveraging the potential of automated valuation models (AVMs) and big data applications in the real estate domain to address the limitations of traditional land valuation methods. The objective of the study is to enhance the transparency and accuracy of land valuation through the integration of a holistic data interpretation system, improved information exchange between AVM projects and property valuation, and the automation of specific workflows for property assessment. The study focuses on the Homagama Divisional Secretarial administrative boundary in Sri Lanka as a case study area for implementing the proposed AVM-based valuation model. The methodology adopts a multi-criteria approach with expert validation, considering various causative factors influencing land value, such as access to transportation, proximity to schools, distance from government and commercial establishments, proximity to hospitals, slope, land use, distance to urban centers, and population. To determine the relative importance of each factor, the Analytical Hierarchical Process (AHP) is utilized, providing a rational and consistent weighting approach. The AVM is developed using ArcGIS software and Weighted Overlay Analysis, effectively capturing the spatial distribution of land values within the study area. The research results in the classification of the Homagama Divisional Secretary area into distinct land value classes, ranging from very low to very high valued areas. In order to ensure the accuracy and validation of the model, the estimated AVM for land lot cadaster data, is reviewed alongside real-world land value data. The valuation map produced presents a graphical representation of the value of each land parcel, thereby facilitating rapid decision-making processes. Additionally, the study highlights the potential for further enriching the model with additional factors beyond those initially considered. In conclusion, this study demonstrates the value of AVMs in revolutionizing land valuation by providing a transparent, accurate, and efficient approach. The proposed model serves as a valuable tool for decision-makers, stakeholders, and property professionals in their efforts to assess land values and make informed real estate investment decisions.