BLOCKCHAIN TECHNOLOGY FOR SECURING INSURANCE DATA AND AUTO-INSURANCE CLAIM

ABSTRACT

Decentralization has gained a lot of attention due to its application
in diverse fields. It is pioneered largely by Bitcoin, a blockchain technol-
ogy and a financial application of decentralization, which has impacted
a lot on how financial transactions happen in a secure manner. The ad-
vantage of using this technology is that there is no central authority to
rely on. Thus a decentralized storage of insurance records would allow
forgeries on the records to be reduced.
We propose a solution to avoid forgery in insurance sector using
Blockchain. The blockchain network in the proposed system will times-
tamp and store insurance data and its associated files in the network
storage. The network is decentralized thus the data is inherently secure.
We also make use of smart contracts available in blockchain tech-
nology to enable auto insurance claim which will reduce a lot of time
wasted in processing the claims. Hence, a system to securely store insur-
ance data using blockchain technology can be implemented or created.

INTRODUCTION

  1. 1 PROBLEM DOMAIN: Traditional health insurance system possesses high possible fraud-ulent or duplicate claims, possible financial loss during digital transac-tions, complex and high time consuming verification processes. This
    is the challenge the insurance industry faces and it possess transactional
    and contractual complexity. The aforementioned problem is an ideal ap-
    plication opportunity for Blockchain implementation. Blockchain can
    provide significant benefits to the private consortium and all of its par-
    ticipants including the insured customers. Blockchain can address the
    fundamental challenges of managing and dragging the distributed digi-
    tal transactions of the problem scenario, as they are extremely secured,
    manageable and high-speed transactions. A lot of work goes into check-
    ing whether a health insurance claim is fraudulent or not. Using the
    consensus property of Blockchain, we can reduce the work that goes
    into checking if a claim is fraudulent or not.
    SCOPE: Now-a-days insurance data are more important, since it can be
    modified by higher authorities. Hence, the security of this insurance
    data becomes questionable. Hospitals may claim insurance for treat-
    ments that were not taken by patients. This happens because we com-
    pletely rely on the centralized record management system of the hospi-tal. So, we need to build a system which records the treatments and bills
    impossible to be counterfeited even by the higher authorities. Hence,
    it becomes impossible to tamper the insurance data of the patient from
    the eye of society. Also it takes lot of time to verify a claim made by
    insurer and to instantiate the transaction. Our main intention is to se-
    cure insurance data using blockchain and to implement auto insurance
    claim. If we implement this in India then the insurance fraudulent will
    be reduced.
  2. RELATED WORKS

2.1 ARTIFICIAL INTELLIGENCE
Automation and AI have transformed almost every sector across the
world, and the insurance industry is no exception. According to Accen-
ture’s Technology Vision for Insurance 2017, 94 percent of “insurance
executives agree that adopting a platform-based business model and en-
gaging in ecosystems with digital partners are critical to their business.”
In 2016, 35 percent of insurers reported over 15 percent in cost sav-
ings from automating systems and processes in the last two years. Au-
tomation of more complex tasks (other than compliance checks or data
entry) such as property assessment and personalized consumer interac-tions over the years has brought frictionless experiences and cut down
redundancy.
Employing AI in the claims process has brought better quality and
lesser time for handling. AI algorithms can save millions lost to fraud-
ulent claims by scouring data and identify errors and trends. The future
is definitely touchless. Machine learning can be useful in evaluating
risk and identifying cross-selling opportunities. Online-only insurance
technology companies, uses AI, machine learning, and big data to “sim-
plify insurance, price risk more finely and distribute cheaply to a mass
market via the internet.” For automated claims processing and property
assessment, PC insurance providers are using drones for more accurate
information and faster processing.

2.2 INTERNET OF THINGS IN INSURANCE
IoT devices, sensors, and telematics have been fast gaining adop-
tion in the insurance sector. Several data streams and sources (wear-
ables, sensors embedded in vehicles, location-based sensors, GIS) cou-
pled with advanced analytics can help insurers improve risk assessment,
price policies based on real data in real time, and proactively encour-
age customers to buy policies for loss prevention. More usage-based
insurance models for connected vehicles and precise actuarial models
are expected with the huge amounts of data (or touchpoints) available
thanks to today’s amazingly connected world. In the auto insurance sec-
tor, for example, the data (speed, time, braking patterns, distance) gives
buyers more say in their premiums; risky driving patterns can serve as
warning signs. Blockchain can be the “network connecting and ordering
data from the multiple devices and apps involved in a multidimensional
process.” (EY, 2016) It can help manage the huge volumes by ensuring
P2P device communication.

Companies such as Aviva and State Farm urge customers to invest
in home sensors (others such as FitSense deal with fit tech to help in-
surers), incentivizing them to help prevent risk to self (e.g. elderly care)
and property. For example, Neos Ventures, UK’s first connected home
insurance specialist, provides preventative smart technology as part of
the policy.
Along with the real-time data and advanced digital capabilities, in-
surers enjoy better customer relationships and risk management, quicker
processing of claims, and selling bundled products. Automating and
streamlining so many data-driven insurance-related processes such as
pricing policies, underwriting, approximating required reserves, and risk
profiling help providers come up with valuable, easy-to-use, and afford-
able products and services.

2.3 BLOCKCHAIN
In the cryptocurrency scenario, the issue between payer and payee
is that there has to be a trusted third party to verify the double-spending
of money and validate the transactions. It leads to a state where all the
participating people has to agree on the trusted authority. The fate of
the entire community is given to a single company that run the trusted
authority. To make all participants of the network to agree upon transac-
tions that were previously made it is important for all the participants to
know all previous transactions. Thus a technology named blockchain is
proposed to store all transactions in a public ledger in a tamper-resistant
manner.
A earlier cryptocurrency called bitcoin used a hard to solve Proof
of Work to validate the transactions. Number of transactions bundled
in a block is hashed in such a way that the hash contains some leading
zeros. The computational complexity increases when number of leadingzeros increases. Thus the whole network is given a CPU based voting
permission.
If the network contains more honest nodes controlling
larger CPU power, then the network will have trusted larger chain of
blocks.

2.4 FINANCIAL APPLICATIONS OF BLOCKCHAIN

2.4.1 Stock Exchange
The stock exchanges list company shares for secondary market to
function securely with trades settling and clearing in a timely manner. It
is now theoretically possible for companies to directly issue the shares
via the blockchain. These shares can then be purchased and sold in a
secondary market that sits on top of the blockchain. NASDAQ Private
Equity has joined hands with a San Francisco based start-up to imple-
ment private equity exchange on top of blockchain.

2.4.2 Asset Management
Assets which can be uniquely identified by one or more identifiers
that are difficult to destroy or replicate can be registered in blockchain.This can be used to verify ownership of an asset and also trace the trans-
action history. Any property (physical or digital such as real estate,
automobiles, physical assets, laptops, other valuables) can potentially
be registered in blockchain and the ownership, transaction history can
be validated by anyone, especially insurers. Everledger is a company
which creates permanent ledger of diamond certification and the trans-
action history of the diamond using blockchain. The verification of dia-
monds can be done by insurance companies, law enforcement agencies,
owners and claimants easily using this blockchain.
2.5 NON-FINANCIAL APPLICATIONS OF BLOCKCHAIN

2.5.1 Health Care
Estonia is implementing a blockchain based health care manage-
ment record to store it in a hacker-proof manner. In that project, logs
of access of health care data and audit data is stored in blockchain. All
the users can see when the data is accessed but access log cannot be
modified.

2.5.2 Music Industry
The process by which music royalties are determined has always
been a convoluted one, but the emergence of the internet has made it
even more complex giving rise to the demand of transparency in the
royalty payments by both artists and song writers. This is where the
blockchain can play a role. The technology can help maintain a com-
prehensive and accurate distributed database of music rights ownership
information in a public ledger. In addition to rights ownership informa-
tion, the royalty split for each work can be determined by smart con-
tracts.

2.5.3 Keyless Security Infrastructure
Namecoin is an alternative blockchain technology that is used to
implement a decentralized version of Domain Name Server. Current
DNS servers are controlled by governments and large corporations, and
could abuse their power to censor, hijack, or spy on a consumer’s inter-
net usage. With blockchain technology internet’s DNS is maintained in
a decentralized manner. Public Key Infrastructure technology is widely
used for centralized distribution and management of digital certificates.
Every device needs to have root certificate of the Certificate Authority
to verify digital signature. While PKI has been widely deployed and
incredibly successful, dependence on a CA makes scalability an issue.
The characteristics of the blockchain can help address some of the limi-
tations of the PKI by using Keyless Security Infrastructure.

2.6 BLOCKCHAIN IN IOT
Blockchain-IoT combination can work perfectly. For once us cases
can be derived as follows.
1. Facilitates the sharing of services and resources leading to the cre-
ation of a marketplace of services between devices.
2. Allows us to automate in a cryptographically verifiable manner
several existing, time-consuming workflows.
There are some practical deployment difficulties in deploying IoT
blockchain. These difficulties range from transactional privacy to the
expected value of the digitized assets traded on the network.
Blockchain-IoT combination can be powerful and can cause signif-
icant transformations across several industries, paving the way for new
business models, novel and distributed application, yet this blockchain
model has some restrictions like scalability. The authors have talked
about minimizing the data copied in a blockchain. They claim that se-curity is not compromised using 51 percent attack yet minimizing the
total data copied among the nodes. A new model of blockchain is pro-
posed for limited memory availability using B language.

REQUIREMENTS ANALYSIS

3.1 FUNCTIONAL REQUIREMENTS
Blockchain provides security for insurance data in this project, the
blockchain should rely on the following requirements.
• Blockchain should not be tampered by third party.
• Blockchain should be stable.

3.2 NON-FUNCTIONAL REQUIREMENTS

3.2.1 Software
• Operating System: Linux
• Programming Language: Python, Go, NodeJS
• Blockchain : Hyperledger Fabric
• Virtual machine : Docker
3.2.2 Performance
The system must be optimized, reliable, consistent and available all
the time.
3.3 CONSTRAINTS
• Hospital bills should be stored in json file format.
• Transaction should be in a specified format and in a specified size
to store that transaction in blockchain.
• No one can retrieve and modify the transaction stored in
blockchain.

4.1 SYSTEM ARCHITECTURE
The block diagram for the entire system is shown in figure 4.1. The
system aims to secure the insurance data by storing it in the blockchain.
Since the insurance data is very large in size, we have restricted our
system to the treatments that falls in the policies alone. Initially the
organization and the peers for each organization is created in the
blockchain network using HyperLedger Fabric. Each organization will
have a certifying authority that endorses the peer that it belongs to its
organization. The Docker is used to virtualize the environment to allow
multiple peers to work on same system. The chaincode is the smart
contract that automatically executes when an transaction takes place.
The data in each block are encrypted using Private Key Encryption and
hashed using SHA256.

4.2 MODULE DESIGN
Our architecture has three phases.
• Pre-processing phase has three modules.

  1. Smart Contracts
  2. Private Key Encryption
  3. Preprocessing
    • Transaction Generation Phase has four modules.
    1. Hashing
    2. Received Packet Verification
    3. Block Validation
    4. Broadcasting
    • Blockchain Miner has three modules.
    1. Data Verification
    2. Genesis Block Creation
    3. Broadcasting Blockchain

4.3 PREPROCESSING PHASE
In this phase we have implemented smart contracts to decide
whether to initiate the transaction or not. We have also encrypted the
data using RSA algorithm to restrict the access.

4.3.1 Smart Contracts
It takes the Bill ID and Bill amount as input. Smart Contracts auto-
matically claims the patient bill amount whose insurance amount must
be less than the insurance limit. And if the insurance amount is less than
or equal to the insurance limit the data is sent to the encrypted phase if
not it is discarded.

4.3.2 Private Key Encryption
It takes the patient treatment details as the input and encrypts the
data using RSA Algorithms and produces the encrypted data.

4.3.3
Preprocessing

It takes the encrypted data as input and checks all the necessary
data to instantiate a claim that has been uploaded or not and produces
the block containing the block header and transaction header.

4.4
TRANSACTION GENERATION PHASE

Once the treatment has been made then patient’s details should be
stored in blockchain. This phase contains hashing the patient details,
treatment details and verification processes. Every transaction begins in
this phase.

4.4.1 Hashing
It takes the block containing the block header and transaction
header as the input and hashes the data in the block using SHA-256
algorithm.

4.4.2 Received Packet Verification
In blockchain every transaction has to be verified by other peers in
a network to take appropriate decision. The following steps has to be
done in this module.

4.4.3 Block Validation
This module is available in every peer of the blockchain network.
The peers in the network obtains the newly mined block from the
miner. They check the validity of the block. If valid they update their
blockchain else discards the new block. Procedure for the block valida-
tion is given below.

4.5 BLOCKCHAIN MINER
In this phase new block will be generated by the miner by solving
proof of work and updates the blockchain. This module contains genesis
block creation, data Verification, broadcasting blockchain.

4.5.1 Data Verification
This module finds a hash below a targeted hash value (PoW). We
have four types of algorithms for consensus in blockchain. Consensus
is the only algorithm we can use. Four types of algorithms are:
• Practical Byzantine Fault Tolerance Algorithm
• Proof-of-work
• Proof-of-Stake
• Delegated Proof-of-Stake

4.5.2 Consensus Algorithm
The proof of work is a computational client puzzle where compu-
tation power is needed to solve the puzzle and the nodes who solve the
puzzle first will add the block to blockchain and will get a reward in
mining network. Once a block is completed, it will be broadcasted to
all the mining nodes and all of them start to mine. The miner node who
solves that first will add that block to block chain.

4.5.3 Genesis Block Creation
In this module, the genesis block for bootstrapping the blockchain
application is created using configtxgen tool which is an inbuilt tool
with Hyper ledger. This tool takes an input file configtx.yaml and out-
puts the binary file which should be used when creating initial setup for
blockchain.

4.5.4 Broadcasting Blockchain
This module broadcasts the newly created block to other peers in
the network.

SYSTEM DEVELOPMENT

In this chapter we discuss the algorithms implemented to develop
this System. We have three phases in our project. They are,
1. Preprocessing Phase
2. Transaction Generation Phase
3. Blockchain Miner

5.1 PROTOTYPE ACROSS THE PHASES
The input and output to each module of the system is described in
this section.

1. Preprocessing Phase
• Smart Contracts It takes the Bill ID and Bill amount as
input.
• Private Key Encryption It takes the patient treatment de-
tails as the input and encrypts the data using RSA Algo-
rithms and produces the encrypted data.
• Preprocessing It takes the encrypted data as input and
checks all the necessary data to instantiate a claim that has
been uploaded or not and produces the block containing the
block header and transaction header.

2. Transaction Generation Phase
• Hashing It takes the block containing the block header and
transaction header as the input and hashes the data in the
block using SHA-256 algorithm.

Received Packet Verification

In blockchain every transac-tion has to be verified by other peers in a network to take appropriate decision. The following steps has to be done in
this module.
• Block Validation This module is available in every peer of
the blockchain network. The peers in the network obtains the
newly mined block from the miner. They check the validity
of the block.

3. Blockchain Miner
• Data Verification This module finds a hash below a targeted
hash value (PoW).
• Genesis Block Creation In this module, the genesis block
for bootstrapping the blockchain application is created using
configtxgen tool which is an inbuilt tool with Hyper ledger.
• Broadcasting Blockchain This module broadcasts the
newly created block to other peers in the network.

5.2 SMART CONTRACT ALGORITHM

5.3 PRIVATE KEY ENCRYPTION
1. Choose two different large random prime numbers p and q
2. Calculate n where n=p*q
3. n is the modulus for the public key and the private keys
4. Calculate the totient (n) where (n)=(p-1)*(q-1)
5. Choose an integer e such that 1¡e¡phi(n) and e is co-prime and e
and (n) share no factors other than 1 gcd(e, (n))=1
6. Calculate d to satisfy the congruence relation de1(mod (n)) i.e.,
de=1+k*(n) for some integer k where d is kept as the private key
exponent

5.4 PREPROCESSING

5.5 HASHING

5.6 CONSENSUS ALGORITHM

5.7 DEPLOYMENT DETAILS
The deployment of the system requires hospital bills should be
available as a json file. We need Hyperledger for blockchain and IDE’s
like NodeJS to deploy the system successfully.

6.1 OUTPUT OBTAINED IN VARIOUS STAGES

6.1.1 Blockchain Network Setup
To implement a blockchain we have to setup network by creating
Orderer and peers. Following figures explain the implementation of
blockchain (6.1 to 6.4)

6.1.2 Genesis Block
The genesis block for bootstrapping the blockchain application
is created using configtxgen tool which is an inbuilt tool with Hyper
ledger. This tool takes an input file configtx.yaml and outputs the binary
file which should be used when creating initial setup for blockchain as
shown in Figure 6.5

6.1.3 Smart Contract
It takes the Bill ID and Bill amount as input. Smart Contracts
automatically claims the patient bill amount whose insurance amount
must be less than the insurance limit. And if the insurance amount is
less than or equal to the insurance limit the data is sent to the encrypted
phase if not it is discarded

6.1.4 Private Key Encryption
It takes the patient treatment details as the input and encrypts the
data using RSA Algorithms and produces the encrypted data as shown
in 6.6

6.1.5 Hashing
It takes the block containing the block header and transaction
header as the input and hashes the data in the block using SHA-256
algorithm.

6.1.6 Preprocessing
It takes the encrypted data as input and checks all the necessary
data to instantiate a claim that has been uploaded or not and produces
the block containing the block header and transaction header. This is
shown in figure 6.7

6.1.7 Tamper Proofness
When the unknown party tries to access the permissioned
blockchain then it will show error and won’t allow him/her to access
the data. This is shown in figure 6.8

6.2 PERFORMANCE EVALUATION
6.2.1 Throughput

Throughput will be measured as the number of successful
transactions per second. A workload can be configured with multiple
clients and threads per clients to saturate the blockchain throughput.

7.1 CONCLUSION
In our system, Hyperledger has been used for building blockchain
which is a permissioned blockchain technology.
The patient details has been considered as a transaction and broadcast in the blockchain network. This triggers smart contract which parses the transaction and gets the insurance policy with regard to the patient details where it
has been generated. This helps to easily find the policy under which
the patient had been covered. find out the spread of heart diseases in
India. This has been implemented using chain code. Blockchain makes
it impossible to tamper these data which removes the control from the
hospital’s central authority. Hence, the insurer or hospital cannot modify
or change the insurance details given.

REFERENCES

[1] Gabriela Ciocarlie, Karim Eldefrawy, and Tancrede Lepoint,
BlockCIS — A Blockchain-based Cyber Insurance System, IEEE
International Conference on Cloud Computing, 2018.
[2] Chuka Oham, Raja Jurdak, Salil S Kanhere, Ali Dorri, and Sanjay
Jha, B-FICA Blockchain based Framework for Auto-insurance
Claim and Adjudication, IEEE International Conference on Cloud
Computing, IEEE, 2018.
[3] Mayank Raikwar, Subhra Mazumdar, Sushmita Ruj, Sourav Sen
Gupta, Anupam Chattopadhyay, and Kwok-Yan Lam, A Blockchain
Framework for Insurance Processes,
9th IFIP International
Conference on New Technologies, Mobility and Security, IEEE,
2018.
[4] Craig Wright and Antoaneta Serguivea, Sustainable Blockchain-
Enabled Services: Smart Contracts, IEEE International Conference
on Big Data, 2017.
[5] Lijing Zhou, Licheng Wang, and Yiru Sun, Blockchain-Based Med-
ical Insurance Storage System, 2018

The College of Engineering, Guindy is a public engineering college in Chennai, India and is India's oldest technical institution, founded in 1794.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store