Abstract: Gypsum, lime and sodium carbonate are selected as additives to eliminate the disbenefits of organic matter on cement stabilization. Response surface methodology (RSM) based on central composite rotatable design (CCRD) is employed for experiment arrangement and optimizing the ratios of additives, in which response values are the unconfined compressive strength values of 7 days and 90 days while the contents of additives are independent variables. Meanwhile, one-factor and mutual-influence analysis are used to investigate the effect of each factor on response values and evaluate mutual-influence among the three independent variables. The results show that strength as response value becomes more sensitive to the content of gypsum and lime as the age increases. However, the situation is reverse for sodium carbonate. With the age of 7 days and 90 days, the mutual-influence for lime and sodium carbonate, gypsum and sodium carbonate are significant respectively. On the premise of humic acid content is 6%, cement content is 15%, the optimal ratios for additives with the age being 7 days and 90 days are presented and once the optimal ratios for additives are applied, the actual strength of stabilized muddy soil can reach 623 kPa(7 d) and 1 213 kPa(90 d).

Key words: humic acid, central composite rotatable design (CCRD), response surface methodology, optimal ratios