A multi-objective optimization approach for disaggregating employment data

Loading...
Thumbnail Image

Date

Authors

Ludick, Chantel Judith
Van Heerden, Quintin

Journal Title

Journal ISSN

Volume Title

Publisher

Wiley

Abstract

In many countries, including South Africa, data on employment is rarely available on a downscaled level, such as building level, and is only available on less detailed levels, such as municipal level. The aim of this research was to develop a methodology to disaggregate the employment data that is available at an aggregate level to a disaggregate, detailed building level. To achieve this, the methodology consisted of two parts. First, a method was established that could be used to prepare a base data set to be used for disaggregating the employment data. Second, a multiobjective optimization approach was used to allocate the number of employment opportunities within a municipality to building level. The algorithm was developed using an Evolutionary Algorithm framework and applied to a case study in a metropolitan municipality in South Africa. The results showed favorable use of multiobjective optimization to disaggregate employment data to building level. By enhancing the detail of employment data, planners, policy makers, modelers and other users of such data can benefit from understanding employment patterns at a much more detailed level and making improved decisions based on disaggregated data and models.

Description

Keywords

Employment data, Multi-objective optimization approach, Disaggregated data and models

Sustainable Development Goals

Citation

Ludick, C. & Van Heerden, Q. 2023, 'A multi-objective optimization approach for disaggregating employment data', Geographical Analysis, doi : 10.1111/gean.12328. NYP.