8C SPS Model: A Seller Product Selection Model for Home-Based Business Sellers Using a Multi-Criteria Decision-Making Approach

Home-based businesses are becoming a necessary tool for sellers of home-based businesses to determine which products they might be able to sell well. Otherwise, most of the sellers still choose products by trial and error, which may bring low demand as well as stock that doesn't sell, resulting in wasted capital. This study proposes an extended Seller Product Selection model from the seller perspective by enlarging the five-criterion benchmark customer-oriented model into eight criteria: Rating, Price, Sold, Discount, Response Chat, Repurchase, Favorite, and Followers. The data utilized in this study are records of food and beverage products from Shopee Indonesia, amounting to 150,000 products. The criteria relevance was tested with Forward Stepwise Multiple Linear Regression, which is additionally supported by the questionnaire given to 30 marketplace sellers online. Regression results for Sold are presented, indicating that all selected criteria were significantly related to the final model, achieving R = 0.953, R Square = 0.908, Adjusted R Squared = 0.908 with a Durbin-Watson of 1.863 (P<.0005) as shown in the table… The ANOVA result also indicates that the total model is statistically significant, F = 206961.417, p < 0.001. Repurchase had the highest statistical correlation with Sold (β = 0.758) among predictors, followed by Favorite (β = 0.299). Questionnaire results illustrate the following: Sold was ranked number one with a mean rank of 2.40 and Borda Weight of 0.183; Repurchase followed closely in position two with a mean rank of 3.63 and Borda Weight of 0.149. In order to validate whether dyads were able to agree on their RfD, Kendall's coefficient of concordance was obtained as follows: W = 0.236 (P < 0001), suggesting a moderate level of agreement among respondents. We conducted product ranking experiments for MOORA, SMART, PSI and TOPSIS under different combinations of criteria and weight assignments based on AHP. The results indicate that the eight-criterion Seller Product Selection model achieves substantially improved ranking stability and consistency compared with the five-state benchmark model presented in the relevant literature. While considering the eight-criterion model, MOORA performed the best in terms of obtaining the lowest MAD value (1,093), MSE value (4,969,891), and the highest Spearman’s Rank Correlation value (0.99868). The results demonstrate that the proposed eight-criterion Seller Product Selection model can be an actionable decision aid to guide home-based business sellers in selecting products that have a better potential in the marketplace using MOORA.

@article{Ginting2026,
  title     = {8C SPS Model: A Seller Product Selection Model for Home-Based Business Sellers Using a Multi-Criteria Decision-Making Approach},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {17},
  number    = {7},
  year      = {2026},
  publisher = {The Science and Information Organization},
  author    = {Selvia Lorena Br Ginting and Zulaiha Ali Othman},
  doi       = {10.14569/IJACSA.2026.0170760},
  url       = {https://doi.org/10.14569/IJACSA.2026.0170760}
}