The electricity supply in Indonesia is considered inefficient, as indicated by the increase in the Cost of Generation Provision (CGP) in the 2016–2021 period. Indications of high supply chain risks are not commensurate with the low demand for LNG. Nevertheless, every power plant on small islands must be served by minimizing costs. This study aims to develop a risk-based small-scale supply chain model to distribute LNG in archipelagic areas by minimizing infrastructure costs, including investment and operations. This paper formulates a risk-based small-scale LNG (SS-LNG) model with milk runs, which are rarely applied in LNG supply chains, thus impacting cost efficiency and the sustainability of LNG supply in Indonesia. The simulation was carried out using the vehicle routing problem with milk run (VRPMR) method. The optimization results show that the 2-vessel scheme is an optimal and efficient risk-based SS-LNG supply chain model. However, it appears that changing the scheme from 1 to 2 vessels generally increases the infrastructure tariff by 3%, or USD 0.11 per MMBTU. This risk-based supply chain model is able to ensure a fuel cost efficiency of 27% for power plants.

In Indonesia, the CoS for power supply increased from Rp. 1025 to Rp. 1334/KWh from 2016 to 2021, respectively; this indicates an inefficient process in electricity provision. One contributing factor to this inefficiency is the existence of many high speed diesel (HSD)-fueled power plants. These are distributed across the Indonesian archipelago with a supply chain that only uses sea transportation. The absence of an economical small-scale LNG (SS-LNG) supply chain also demonstrates the inadequate infrastructure for distributing LNG to refineries. This study aims to analyze the probability of risks that occur in SS-LNG supply chains in the Indonesian archipelago. The analytical methods used are descriptive statistical analysis and Delphi analysis through in-depth interviews and Focus Group Discussions (FGD) with experts. Results showed that the SS-LNG supply chain process in Indonesia includes LNG loading, unloading, shipping, picking, storage, regasification, and distribution. There are 30 risk indicators that may occur, with the highest risks including ship accidents, equipment damage, lack of transport ships, bad weather, earthquakes, tsunami, poor safety cultures, and low levels of safety leadership. These risk indicators can be used in implementing SS-LNG.