Innovative Sudoku Designs with Rectangular Sub-zones: Enhancing Experimental Precision in Paddy Field Trials

Abstract

This study investigates the application of a Sudoku square design with rectangular subzones for a field experiment on paddy cultivation conducted in 2022-2023 in West Bengal, India. Traditional row column experimental designs, such as the Latin Square Design (LSD), have been widely used in agricultural research. However, these designs have limitations in estimating a variety of effects due to their structure. The research addresses this limitation by employing a sudoku square design, which has introduced a notable framework for experimental analysis. Four distinct statistical models were employed to analyze the yield data. The models captured varying sources of variation. Type 1 identified three: row block, column block, and subzone. Type 2 added rows within BR and columns within BC. Type 3 included H-square within RB and V-square within CB. Type 4 combined all, revealing six sources in total. Although not all additional sources were statistically significant, variations such as treatment, row, subzone, row block, H-square within RB, and V-square within CB showed statistical significance. Model demonstrated the enhancement of their analytical capability to estimate the effects of these additional sources. The varying number of additional sources identified by each model highlighted the flexibility and robustness of the Sudoku square design in capturing complex effects in the experiment.