Municipal waste logistics forecasting -- Jakarta Smart City
Jakarta's solid waste management agency (DINAS LINGKUNGAN HIDUP) operates waste collection logistics for a city of 10M+ residents across hundreds of collection points and sub-districts. Route scheduling was reactive -- trucks were dispatched on fixed schedules irrespective of actual waste accumulation, resulting in both under-collection (overflow complaints) and over-collection (idle truck capacity wasted on empty routes), with cascading effects on fuel cost and emissions.
Waste volume forecasting at the collection-point grain is a noisy time series problem with strong but irregular seasonality: public holidays, cultural events (Lebaran, Christmas), local market days, and weather events (heavy rain flooding) all create non-stationary spikes that additive decomposition models struggle to capture. A single city-level forecast is insufficient -- route optimization requires district-level and collection-point-level granularity, multiplying the forecasting problem by hundreds of independent series with varying data quality.
Developed ensemble time-series forecasting pipeline: Facebook Prophet (primary, for interpretable seasonal component estimation and holiday handling) benchmarked against ARIMA, SARIMA, and exponential smoothing baselines. LSTM explored for capturing non-linear temporal dependencies in high-density districts. Exogenous variables integrated: public holiday calendars, weather data (rainfall, temperature), and local event schedules.
Conducted pre-post causal analysis to measure the impact of Jakarta's 2020 plastic bag ban on waste complaint volume using citizen report data from JAKI and Qlue platforms (100,000+ complaint records). Applied difference-in-differences methodology controlling for seasonal and macroeconomic confounders (COVID-19 mobility restrictions required explicit control). Results provided statistically significant evidence of complaint reduction post-intervention.
Built Tableau dashboards for spatial-temporal analysis of waste volume patterns by district, translated into operational route recommendations for logistics planners. Results and policy findings presented to 500+ attendees at ICISS 2021 (IEEE-sponsored international conference).
15% operational efficiency improvement in waste collection logistics across a 10M+ resident urban grid -- translating to direct fuel cost reduction and emissions savings. Policy causal analysis provided empirical evidence for the plastic bag ban's effectiveness, establishing a methodology applicable to evaluating future environmental regulations at city scale. First-authored IEEE conference paper presented internationally.