Published Paper Details:

Supply chain management (SCM) architectures require high security & high quality of service (QoS) in order to effectively track &record product-related information. A wide variety of SCM models are proposed by researchers to perform this task. These models use encryption standards like elliptic curve cryptography (ECC), hashing, key-exchange mechanisms, etc. in order to improve system security. Due to which quality of service (QoS) parameters like retrieval delay, throughput, and energy efficiency are reduced. To maintain a good level of QoS, various machine learning optimizations are also proposed by researchers. It is observed that blockchain-based models are capable of achieving both these constraints for small to medium scale SCM architectures. This is due to their immutability, traceability, transparency, and distributed computing capabilities. Due to immutability, entries stored on the chain cannot be modified, thus all SCM transactions are tamper proof. While, due to distributed computing, the model is capable of adding blocks from different entities. Thus, making blockchains a suitable choice for a wide variety of SCM application deployments.But as length of blockchain increases, this performance is exponentially reduced, which limits scalability of blockchain-based SCM architectures. To improve this scalability, a novel delay-aware sidechaining model is proposed in this text. The proposed model is capable of scaling existing blockchain SCM architectures and create delay-aware sidechains which are aggregated using a Genetic Algorithm (GA) approach. The proposed model was tested on a wide variety of SCM deployments, and an average delay reduction of 23% was observed when compared with single blockchain-based implementations. Furthermore, the computational complexity was also observed to be 15% lower when compared with single blockchain, which makes the proposed SCM model highly scalable. The reduction in computational complexity is due to reduced reading & verification delays, which reduces hash evaluation effort, thereby improving overall SCM performance for various deployment scenarios.