Estimating performance in relation to the expectation is a key component of many machine learning algorithms for decision-making. Measuring performance in accordance with expectations may not be very useful in many real-world situations. In this article, with deployment to a public dataset, we examine the viability and comparative analysis of Deep Learning techniques to anticipating the demand problem. We compare Deep Learning performance to that of various model approaches, such as Random Forest, Gradient Boosted Trees, and Support Vector Machine, using RMSE performance criteria. The forecasting issue is crucial for organizational decision-making. When making strategic decisions on valuable resources, riskaverse goals should be taken into account. This article aims to demonstrate the usage of ML models for forecasting and decision making to improve performance of data analysis of an organization. And to demonstrate that, especially when decision-makers are dealing with complicated limitations data, a Deep Learning algorithm can be a dominant answer to Machine Learning challenges for forecasting and decision-making.