Pharmaceutical products are subject to a high degree of demand variation in the industrial market. We have seen that the global spread of the COVID-19 pandemic has had a devastating impact on such supply chains. In this paper, we describe the integrated procedure for market demand forecasting and purchase order generation in the pharmaceutical demands on a national level. Thereafter, we process and analyze data on international flows of labeled pharmaceutical goods to extract knowledge and forecast. We undertake cross-national analysis to forecast demand for over 240 countries around the world. In this study, we rely on an OEC data source "The Observatory of Economic Complexity" to visualize, understand and interact with the historic international trade data. The objective of this paper is to predict future demands from the quantity of previous features by considering the effects of external factors by employing several existing machine learning approaches applied on pharmaceutical import/export data. Forecasting scenarios for demand calculations such as LSTM, CNN, multiple linear regressions and other models are tested. The given application illustrates the effectiveness of these approaches. Their performance measures in both cross validation and overfitting learning modes are studied experimentally, and their practical implications are discussed.