Machine learning PODA model projects the impact of COVID-19 on US motor gasoline demand – Green Car Congress

A team from Oak Ridge National Laboratory (ORNL), Aramco Services Company, MIT, the Michigan Department of Transportation and Argonne National Laboratory has developed a machine-learning-based model (Pandemic Oil Demand Analysis, PODA) to project the US medium-term gasoline demand in the context of the COVID-19 pandemic and to study the impact of government intervention. Their open-access paper appears in the journal Nature Energy.

The PODA model is a machine-learning-based model to project the US gasoline demand using COVID-19 pandemic data, government policies and demographic information. The Mobility Dynamic Index Forecast Module identifies the changes in travel mobility caused by the evolution of the COVID-19 pandemic and government orders. The Motor Gasoline Demand Estimation Module quantifies motor gasoline demands due to the changes in travel mobility. Ou et al.

They found that under the reference infection scenario, US gasoline demand grows slowly after a quick rebound in May, and is unlikely to recover to a non-pandemic level prior to October 2020.

Under both the reference and a pessimistic scenario, continual lockdown (no reopening) could worsen the motor gasoline demand temporarily, but it helps the demand recover to a normal level more quickly due to its impact on infection rate.

Under the optimistic infection scenario, the projected trend of motor gasoline demand will recover to about 95% of the non-pandemic gasoline level (almost fully recover) by late September 2020.

However, under the pessimistic infection scenario, a second wave of infections in mid-June to August could lower the gasoline demand once morebut not worse than it was in April 2020.

The researchers conclude that their results imply that government intervention does impact the infection rate, which thereby impacts mobility and fuel demand.

Projections of the evolution of COVID-19 pandemic trends show that lockdowns help to reduce COVID-19 transmissions by as much as 90% compared with the baseline without any social distancing in Austin, Texas. However, this unprecedented phenomenon could last for a few years: Kissler et al. suggested that, even after the pandemic peaked, COVID-19 surveillance should be continued as a resurgence in contagion could be possible as late as 2024. Therefore, beyond the immediate economic responses, the longer-term impact on the US economy may persist well beyond 2020. An effective forecast or estimate of the pandemic impacts could help people to well prepare and navigate around unknown risks. More specifically, reliably projecting the oil demand, a critical leading indicator of the state of the US economy, is beneficial to related business activities and investment decisions.

There are studies that discuss the impacts of unexpected natural hazards and/or disasters on energy demand and/or consumption and studies that evaluate the impacts of previous pandemics on tourism and economics . However, few studies have quantified and forecast the oil demands under multiple pandemic scenarios, and this research is desperately needed.

To date, studies focused on the energy impacts of the COVID-19 pandemic are limited to the short-term energy outlook released by the US Energy Information Administration (EIA); this outlook uses a simplified evolution of the COVID-19 pandemic to forecast the US gross domestic product, energy supplies, demands and prices until the fourth quarter of 202115. In this work, we develop a model that combines personal mobility with motor gasoline demand and uses a neural network to correlate personal mobility with the evolution of the COVID-19 pandemic, government policies and demographic information.

Ou et al.

The model contains two major modules: a Mobility Dynamic Index Forecast Module and a Motor Gasoline Demand Estimation Module. The Mobility Dynamic Index Forecast Module identifies the changes in travel mobility caused by the evolution of the COVID-19 pandemic and government orders, and it projects the changes in travel mobility indices relative to the pre-COVID-19 period in the United State.

The change in travel mobility, which affects the frequency of human contact or the level of social distancing, can reciprocally impact the evolution of the pandemic to some extent.

The Motor Gasoline Demand Estimation Module estimates vehicle miles traveled on pandemic days while it considers the dynamic indices of travel mobility, and it quantifies motor gasoline demands by coupling the gasoline demands and vehicle miles travelled.

The neural network model, which is the core of the PODA model, has 42 inputs, 2 layers and 25 hidden nodes for each layer, with rectified linear units as the activation function. In the PODA model, the potential induced travel demand due to the lower oil prices under the COVID-19 pandemic is not explicitly considered.

Resources

Ou, S., He, X., Ji, W. et al. (2020) Machine learning model to project the impact of COVID-19 on US motor gasoline demand. Nat Energy doi: 10.1038/s41560-020-0662-1

See the rest here:
Machine learning PODA model projects the impact of COVID-19 on US motor gasoline demand - Green Car Congress