The Rise of Artificial Intelligence in Insurance – …...• The rise of Artificial Intelligence has brought in technology transformation in many industries • This presentation

Boyi Xie | NAIC Webinar | 09/ 26/ 2018

The Rise of Artificial Intelligence in Insurance – Applications Boyi Xie , Swiss Re

NAIC Webinar | September 2 6 th, 2 018

Boyi Xie | NAIC Webinar | 0 9 / 2 6 / 2 0 1 8


• The rise of Artificial Inte lligence has brought in technology transformation in many industries

• This presentation will show use cases in insurance that leverage the recent developments of AI

• These use cases are being used across the entire insurance value chain

• Artificial Inte lligence is used here as a broad terminology that refers to the modeling methods, tools and platforms, applications, big data utilization, and the thought process for problem solving in the fie ld of Artificial Inte lligence

Overview

2

Sales & Marketing Application / Submission

Costing & Pricing Underwriting Inforce

Management Claim

Management Risk

Management


• Use big data and machine learning to better identify and analyze client needs

• Cluster clients based on the need of insurance coverage

Prospecting Clients – Filling the Protection Gap

3

Sales & Marketing Application / Submiss ion


Management Claim

Management Risk

Management

Hail Insurance

Wildfire Insurance

Flood Insurance

Auto Insurance Life Insurance

An analogy to product recommendation or

movie recommendation.


• Monitor market movements to identify business opportunities and mitigate risks

Originating New Business – Market Intelligence

4



Management Claim

Management Risk

Management

M&A

Credit

Management

Capital

Share Price

EPS

P/ E

RBC

Curr Ratio

Debt/ Equity

Solvency

Liquidity

Credit Risk

Volatility

VaR

Lower Level Signals Intermediate Level

Signals Higher Level

Signals

Attributes extracted from unstructured (text) and s tructured data from different data sources.

Y/ N

Business Opportunity


Expediting Application Process

5



Management Claim

Management Risk

Management

• Expedite the application process while maintaining an accurate quantification of risk

Important predictors

Unimportant predictors

Reduced claim amounts per policy

Increased claim amounts per policy

Lengthy Applicat ion Forms

Expedited Applicat ion Process

Efficiency & Insights to write better and more

business

1 2

3

Better Estimation of Risk

Expediting Applications

Write More Business


• Automate the information extraction from submission documents to improve efficiency

Streamlining Submission Handling

6



Management Claim

Management Risk

Management


• Algorithms developed in AI enrich the availability of s tatis tical models for loss estimation

• These models can bring in more accurate loss prediction

Improving Loss Estimate

7



Management Claim

Management Risk

Management

Total incurred loss

Number of losses

Claim Rate

Claim Frequency

Average amount incurred per loss Claim Severity

Predictive modelling targets

Rich availability of statistical models

o Linear Regression o Logistic Regression o Generalized Linear Models o Decision Tree o Random Forest o Gradient Boosting Machine o Support Vector Machine o Bayesian Models o Deep Learning

𝑅𝑅𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 𝜃𝜃 = 𝑅𝑅𝑟𝑟𝑒𝑒𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟 𝜃𝜃 + 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝜃𝜃

=1𝑁𝑁� 𝐿𝐿 𝑃𝑃𝑟𝑟 ,𝑓𝑓 𝑥𝑥𝑟𝑟;𝜃𝜃 +

𝜆𝜆2𝑁𝑁

| 𝜃𝜃 |2𝑁𝑁

𝑟𝑟=1


• The effectiveness of the underwriting rules/ triggers for loss estimate can be evaluated

• A better underwriting ecosystem can be built to improve the accuracy and efficiency of underwriting

Accelerating Underwriting

8



Management Claim

Management Risk

Management

Underwriting Triggers

Manual Underwriting Underwriting

Decision

Automatic Underwriting – e liminating unnecessary UW triggers can improve efficiency of the UW process


• Machine learning models can help better evaluate the risk level during underwriting

Classifying the Risk Class

9



Management Claim

Management Risk

Management


• Use inforce data to estimate the likelihood of a policy lapse , which helps proactive customer engagement

Predicting Lapse

10



Management Claim

Management Risk

Management

Use algorithms to understand the behavior of policyholders

Insurance Product

Policy Terms

Age

Premium Payment History

Health Status

Life Changing Events

Lapse? Y/ N or

P(Lapse)


• Use of sate llite images to expedite damage estimation

• Improved claim estimation

• Reserving/ capital preparation

• Proactive customer care

Estimating Claims

11



Management Claim

Management Risk

Management

𝒑𝒑𝒊𝒊 𝒑𝒑𝒊𝒊 = f( - )

Damagibility 𝐷𝐷 Sate llite 𝑆𝑆𝑟𝑟𝑎𝑎𝑎𝑎𝑟𝑟𝑟𝑟 Sate llite 𝑆𝑆𝑏𝑏𝑟𝑟𝑎𝑎𝑏𝑏𝑟𝑟𝑟𝑟


• Mining large scale of data to screen risks, and identify trends of risk development

Identifying Potential Risks

12



Management Claim

Management Risk

Management

News Articles

Machine Learning Based Classification

Document Retrieval

Content Representation

Machine Learning-Based

Class ifie r Topic

Quantification

Feedforward Neural

Network

Regular Expression & Fuzzy Query

Word and Document Embedding

Institution Actions

Travel Restrictions

Shortage of Medical Supply

Antiviral

Unexpected Death


• The rise of Artificial Inte lligence has brought in technology transformation in many industries

• This presentation discussed use cases in insurance that leverage the recent developments of AI

• Insurance is a highly specialized industry and would need technology customization

• It is s till an early s tage of the application of AI in insurance but we have already seen many opportunities

• While there are a lot of excitement, we would s till need to proceed with cautious in the AI adoption

Summary

13

Sales & Marketing Application / Submission


Management Claim

Management Risk

Management

Boyi Xie | NAIC Webinar | 09/ 26/ 2018 14

Boyi Xie | NAIC Webinar | 09/ 26/ 2018 Boyi Xie | NAIC Webinar | 0 9 / 2 6 / 2 0 1 8

Legal notice

15

©2018 Swiss Re. All rights reserved. You are not permitted to create any modifications or derivative works of this presentation or to use it for commercial or other public purposes without the prior written permission of Swiss Re.

The information and opinions contained in the presentation are provided as at the date of the presentation and are subject to change without notice. Although the information used was taken from reliable sources, Swiss Re does not accept any responsibility for the accuracy or comprehensiveness of the details given. All liability for the accuracy and completeness thereof or for any damage or loss resulting from the use of the information contained in this presentation is expressly excluded. Under no circumstances shall Swiss Re or its Group companies be liable for any financial or consequential loss relating to this presentation.

Documents

The Rise of Artificial Intelligence in Insurance – …...• The rise of Artificial Intelligence has brought in technology transformation in many industries • This presentation