An Intelligent Context-Aware Management Framework for Cold Chain Logistics Distribution

Contextual information can accurately describe the user's time, location, environment, activities, the use of devices, and so on. The recommender systems with contextual information have better prediction accuracy, coverage, and user satisfaction. However, the application of conventional context-aware framework in some special fields has some drawbacks. For example, in the field of cold chain logistics distribution, the conventional context-aware framework cannot adapt to the multi-dimensional, complex and dynamic contexts of cold chain logistics, the lack of context-aware risk management, and the lack of traceability of cargo contexts. In this paper, we propose an intelligent context-aware management framework for cold chain logistics (CMFCCL) distribution that contains acquisition framework, recommender systems framework, risk management framework, tracing back framework, and user portrait framework of cold chain logistics distribution. Cold chain info data set is used for simulation, and three other recommendation approaches are considered in comparison. The experimental results indicate that CMFCCL can reduce the mean absolute error and root mean squared error of the data set by about 8.37-19.99. Thus, the evaluation presents encouraging results, indicating that CMFCCL would be useful in the context-aware management of cold chain logistics distribution.