multi-objective design of a blood supply chain based on sustainability approach and demand prediction using deep learning algorithm

One of the most critical components of a healthcare system is the blood supply chain, which accounts for a significant proportion of the system's expenditure. therefore, any improvement in the blood supply chain's performance can significantly increase healthcare systems' efficiency and cost-effectiveness. the main challenge in managing blood products lies in supply and demand uncertainty, leading to a trade-off between scarcity and waste, especially in developing countries. in addition, the predictive power of deep learning models for estimating and forecasting the demand for blood products has yet to be sufficiently explored. this paper proposes a multi-objective model to optimize the blood supply chain network. the objectives include minimizing blood delivery times, reducing economic costs in the supply chain, reducing carbon dioxide emissions, and maximizing demand satisfaction as an aspect of social sustainability. given the uncertainty of blood supply and demand, a deep learning model based on the cnn method is used to predict blood demand. the lp-metric algorithm is used to solve the model in the gams software, and the sa simulation algorithm is used to validate the results. the calculation results show that the sa algorithm performs better in optimizing the first objective function, resulting in a shorter product delivery time. however, the lp-metric method performs better for the second and third objective functions.