Supply Chain Shock

Preparing for disruptions from natural disasters

Overview

• This project aims to develop the Supply Chain Shock Model to address business failures following natural disasters. Collaborating with Prof. Doyne Farmer from the University of Oxford, the model uses multi-agent modeling to measure indirect supply chain risks. Outcomes include its use in business growth strategies, insurance pricing, risk management advice, and assessing risks in portfolio companies. The approach involves building supply chain data from expert surveys and existing disaster modeling software. Initial research will focus on sector-level analysis, progressing to firm-level analysis after verifying its feasibility.

​

Detail

Background

• Natural disasters can indirectly impact businesses by disrupting their supply chains, even if the companies themselves aren't directly hit. This leads to significant societal issues, as studies in the U.S. show that 40% of businesses don't reopen after a major natural disaster. The Supply Chain Shock Model aims to quantitatively estimate these indirect risks.

​

Hypothesis

• The model's outcomes could assist in:

  • Business expansion decisions and pricing for insurance coverage against profit loss and business continuity expenses.

  • Providing risk management consultation for businesses facing natural disasters.

  • Conducting risk assessments for investment portfolio companies.

​

R&D

• This research is a joint effort with Prof. Doyne Farmer from the University of Oxford and his company, Macrocosm. Prof. Farmer's research focuses on economics, including agent-based modeling, financial instability, and technological progress. 

• The model uses multi-agent modeling, representing each company as an independent entity to simulate supply chain impacts. 

• Supply chain data will be compiled from external sources and industry expert surveys, utilizing established disaster modeling software commonly used by the MS&AD group to estimate disaster scales.

​

Next Step

• The project will start with a simpler industry (sector) level analysis, expected to take about six months. Upon successful feasibility verification, it will advance to a more detailed firm-level analysis. This progression allows for thorough testing and refinement of the model before applying it to specific companies.