Hybrid SEM-neural networks for predicting electronics logistics information system adoption in Thailand healthcare supply chain

The aim of this work is to examine the adoption of the electronics logistics information system in healthcare industry in Thailand by using structural equation modelling (SEM) approach. Neural network is then employed to test and confirm the research model. These approaches are applied to analyse the effect of all independent constructs and behavioural intention to adopt e-logistics information system by healthcare workers. Unified theory of acceptance and use of technology 2 (UTAUT2) was used to examine electronics logistics information system adoption in the hospitals. Confirmatory factor analysis (CFA) was applied to determine how well the measured variables represent the constructs. SEM was then introduced to analyse the relationship among the variables. Lastly, neural network was applied to predict the relative importance of each independent variable. The study from SEM revealed that seven potential variables of behavioural intention from UTAUT2 for the adoption of e-logistics can be compressed into six variables (performance expectancy, perceived value and support, price value, social influence and facilitating conditions, perceived ease of use and habit). Three significant variables for the e-logistics in hospital adoption in Thailand (performance expectancy, effort expectancy, and habit) are proven to be statistically significant.