Purpose
The objective of this research work is to design a data-based solution for administering traffic organization in a smart city by using the machine learning algorithm.

Design/methodology/approach
A machine learning framework for managing traffic infrastructure and air pollution in urban centers relies on a predictive analytics model. The model makes use of transportation data to predict traffic patterns based on the information gathered from numerous sources within the city. It can be promoted for strategic planning determination. The data features volume and calendar variables, including hours of the day, week and month. These variables are leveraged to identify time series-based seasonal patterns in the data. To achieve accurate traffic volume forecasting, the long short-term memory (LSTM) method is recommended.

Findings
The study has produced a model that is appropriate for the transportation sector in the city and other innovative urban applications. The findings indicate that the implementation of smart transportation systems enhances transportation and has a positive impact on air quality. The study's results are explored and connected to practical applications in the areas of air pollution control and smart transportation.

Originality/value
The present paper has created the machine learning framework for the transportation sector of smart cities that achieves a reasonable level of accuracy. Additionally, the paper examines the effects of smart transportation on both the environment and supply chain.