Integration of neural network and AP-NDEA model for performance evaluation of sustainable pharmaceutical supply chain

Desirable performance of sustainable pharmaceutical supply chain plays a key role in health attainment and performance evaluation is an essential element of effective pharmaceutical supply chain. Several models have been developed for performance evaluation of supply chains. The important point is that the model should be comprehensive and produces the reliable results. For this purpose, comprehensive criteria for evaluation of all levels at the supply chain is identified based on the revised perspectives of Balanced Scorecard. Considering the network nature of the supply chain, Anderson Peterson Network Data Envelopment Analysis (AP-NDEA) model is used to measure efficiency and rank efficient units. To overcome the weakness of this model, this paper for the first time integrates the predictive Neural Network with the AP-NDEA model called Neuro-AP-NDEA. The proposed model estimates the efficiency measurement function in the shortest time, results in computational savings in memory and is more resistant to statistical disturbances. To make the evaluation model more effective and realistic, Interval Evidential Reasoning with linguistic Interval Fuzzy Belief degree (IFB-IER approach) is applied. A numerical example is provided to illustrate the model. The analytical results indicate that the Neuro-AP-NDEA model allows for an accurate prediction and more efficient performance evaluation than the AP-NDEA model.