Urban Logic
[03a]
Urban Simulation
[Desciption]
A demographic forecasting model projecting Riyadh's population by age gender and district over a 50-year horizon. Implements Cohort Component Modelling extended into a household-based agent framework for district-level resolution across three growth scenarios.
[Model Objective]
A demographic forecasting model that projects Riyadh's population by age, gender, and district over a 50-year horizon (2020–2070). It implements the Cohort Component Method (CCM) — tracking births, deaths, and migration across 5-year time steps — and extends it into a household-based agent framework for district-level resolution. Three growth scenarios (high, mid, low) reflect uncertainty in fertility and migration trends, allowing planners to stress-test provision decisions against a range of demographic futures.
PhD Research
2023
Population Growth & Forecasting Model
Know more
The model monitors total population, school-age cohort size by stage and gender, and district-level demographic profiles across all 50-year projection steps. Validated against Riyadh's 2016 ADA estimates, the CCM and ABM outputs show strong agreement — with CCM producing smoother growth curves and the ABM capturing district-level fluctuation. School-age population estimates carry an average error of 11% MAPE, with the adjusted population baseline suggesting a 7% upward revision in required school capacity relative to original estimates.
[Model Components]
CCM engine projecting population by 5-year age groups and gender · Fertility, mortality, and internal migration rate modules · Household-based population synthesiser for district-level disaggregation · School-age population extractor by education stage (elementary, middle, secondary) and gender · Three-scenario design with sensitivity testing
[Model Results]
The model monitors total population, school-age cohort size by stage and gender, and district-level demographic profiles across all 50-year projection steps. Validated against Riyadh's 2016 ADA estimates, the CCM and ABM outputs show strong agreement — with CCM producing smoother growth curves and the ABM capturing district-level fluctuation. School-age population estimates carry an average error of 11% MAPE, with the adjusted population baseline suggesting a 7% upward revision in required school capacity relative to original estimates.
[Model Components]
Cohort Component Modelling (CCM) · ARIMA Extrapolation · BATS Forecasting · Age-Specific Fertility & Survival Rates · Household-Based Population Synthesis · Multi-Scenario Design (High / Mid / Low growth)
[Tags]
["CCM", "Demographic Forecasting", "ABM", "Population Synthesis", "AnyLogic"]