Optimization of bundling strategy decisions and inventory allocation with the integration of the degree of contingency and dead stock levels in a multiple time period setting

• Main Authors: Franco, Edward John A., Santos, Mikhaela Carissa C., Suyom, Denise Ericka M.
• Format: text
• Language: English
• Published: Animo Repository 2017
• Subjects: Bundling (Marketing); Marketing; Operations Research, Systems Engineering and Industrial Engineering
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/18679
• Institution: De La Salle University

Product bundling is a marketing approach that practices the selling of two or more products in the market in efforts to make more sales. This practice is common in various industries, from manufacturing up to the service industry. These industries have continued to use product bundling as a marketing strategy in order to attract more consumers into buying their products, as well as, to replenish their products in the inventory. In a manufacturing setting, products are bundled in either by pairs or in groups. There are three current bundling strategies that are being practiced in the market, which are namely: Pure component, pure bundling, and mixed bundling. The nonlinear programming model aims to optimize these strategies while also considering the selection of components for the bundles. The proposed model integrates the degree of contingency, which measures the complementarity between products, and the dead stock component of the product which would be able to tackle the purpose of product bundling to push out slow to non-moving inventory. The model used is a multiple integer non-linear programming model, which was run through the GAMS Optimization tool. The significant constraints that were used includes customer heterogeneity wherein a high customer heterogeneity would lead to a higher probability of product bundling while a lower customer heterogeneity would lead to a higher probability of being sold as a pure component. Another significant constraint includes integrating the degree of contingency and the dead stock features wherein a fit value was used as a parameter in order for the model to decide which of these features used would be prioritized. The proposed model is validated through different scenarios with varying incremental input parameters in order to determine the effects of the variables to the objective function and the sensitive points in the model. For the sensitivity analysis, design of experiments through design expert was used to analyse effects of single and combination parameters to decision variables. In the sensitivity analysis, the model was able to show the significance of being able to integrate the factors of degree of contingency and dead stock products in identifying which products to bundle. The results were able to show that products with a high degree of contingency and high dead stock components would produce the most profit for the firm. The possibility however of bundles occurring with low degree of contingency and with low dead stock component is still present which gives the importance of being able to integrate both features in selecting which products to bundle. The importance of integrating the component selection and the different bundling strategies gave the model the flexibility of changing bundling strategies through time, which would improve the profitability of the firm through time as well.