Development of a smart map for planning and evaluating urban infrastructure efficiency based on human activity models in an urban environment

The modern world, including Kazakhstan, is already inseparable from information technology. The volume of data generated by people as a result of their activity in the IT world is steadily growing and accumulating. This big data has potential and carries information that can be extracted and used in practice to improve efficiency in a wide variety of fields, such as education, urban logistics and business. Thanks to recent advances in science, we are able to extract information from data that could be useful for urban infrastructure planning and for developing the smart city concept.

Project Objectives

Developing a system for planning and evaluating the efficiency of urban infrastructure processes depends heavily on the quantity and quality of digital data. Therefore, the key objectives of this project are:

• to search for and collect potentially useful data on human activity in an urban environment;
• to define the final list of models for building a smart map with statistical data. Depending on the data available, the list may include, for example, models for identifying overcrowded bus and trolleybus routes, models for identifying the most effective (in terms of sales) business locations, and models for identifying locations for building or opening educational institutions, Public Service Centers and other urban infrastructure facilities;
• to develop machine learning models, which in turn includes formulating hypotheses, bringing unstructured data into a defined structured form and improving model quality metrics;
• to prepare and configure the technical server infrastructure on which statistical and machine learning models will be built and data arrays processed and stored;
• to process and store big data, which includes configuring data sources to load into a single repository and setting up ETL processes for the production use of planning and optimization models;
• finally, to combine the results of the developed machine learning models into a single smart map visualizing statistics across the city.

Novelty

• in the use of mathematical and statistical approaches for planning the operation of urban infrastructure in Kazakhstan;
• in the development of a smart map (a visual tool) that will combine the results of models of different profiles;
• in the development of a decision support system for urban development in Kazakhstan under conditions of high urbanization based on machine learning models;
• in the development of the Smart City concept in Kazakhstan. The research results will be able to address a number of tasks in planning and evaluating the efficiency of urban infrastructure processes using a combination of statistical and machine learning methods [6]. The result of the project will be a smart map that will help evaluate the efficiency of the existing infrastructure of a megacity and plan the operation of future infrastructure. The developed set of models and decision support systems will potentially lead to the following results:
• Kazakhstan will have its own intelligent system for planning and optimizing urban infrastructure processes based on models analyzing human activity in a megacity.
• The developed system will give impetus to the development of smart systems under the Smart City concept in Kazakhstan.
• The developed system will help optimize processes in the existing infrastructure of a megacity and plan the operation of future infrastructure.

• Copyright Certificate: "Advanced" tool facilitating high-quality and effective analysis of (un)structured data, No. 13562 dated November 27, 2020.