Predictive models are used to guess values of a variable of interest based on values of other variables. Developments in mathematical foundations of predictive models were boosted by increasing computational power of personal computers and availability of large datasets in the era of "big data" that the people have entered. It is now becoming clear that the people have got to control the models and algorithms that may affect the reader. In this chapter, the authors discuss the Local-Interpretable Model-agnostic Explanations method. They describe local-stability plots that are useful to investigate the sources of a poor prediction for a particular single observation. The authors mainly focus on the impact of the model exploration and explanation tools rather than on selected methods. They believe that by providing the knowledge about the potential of model exploration methods and about the language of model explanation, the authors help the reader in improving the process of data modelling.