Hybrid Approach: A Multi-Criteria Decision-Making Method for Reverse Logistics Provider Selection

2025-04-24T07:52:00+00:00

Due to governmental laws and increased environmental consciousness among people aimed at reducing waste, the rise in the return of old goods has gained relevance as a logistics problem. Industry-specific infrastructure, monitoring information systems, and return-handling equipment are often required. Since they specialize in reverse logistics, third-party providers (3PRLPs) are currently in high demand across a wide range of businesses.

The Hybrid decision-making method is a novel and methodical approach to decision-making that aids people or organizations in assessing a group of possibilities in light of several factors. Therefore, in this study, the third-party providers (3PRLPs) who specialize in reverse logistics were ranked utilizing Hybrid decision-making method methodologies.

This study offers a novel method for identifying the best 3PRLP (third-party reverse logistics provider), using a multi-criteria group decision-making (MCGDM) model within the Hybrid decision-making method framework. The study digs into the relationships between the factors considered during this selection process, which finally results in the selection of the best 3PRLP out of a possible pool of six options. The Hybrid decision-making method technique is used in this research to prioritize orders based on how closely they resemble the ideal solution. The practical utility of the model is demonstrated through a case study focused on the battery manufacturing industry in India.

The 3PRLP3 got the first rank and the 3PRLP2 got the last rank. 3PRLP1 is second rank, 3PRLP5 is third rank, 3PRLP6 is fourth rank and 3PRLP4 is fifth rank.

AI-Driven Risk-Adaptive Authorization for Multi-Tenant Cloud APIs - Microsoft Azure

2025-06-13T03:08:32+00:00

Application Programming Interfaces (APIs) have become the nervous system of modern financial-services platforms; yet fractured authorization logic remains a dominant breach vector, exposing organizations to data exfiltration, fraudulent transactions and regulatory fines. We introduce a cloud-native Contextual Authorization Framework (CAF) embedded in an enterprise Shared Services Platform (SSP) that supports more than 200 customers facing and internal micro-services worldwide. CAF sits behind Azure API Management, authenticates callers via OAuth 2.0 / OpenID Connect, and merges an ensemble machine learning risk score—Isolation Forest, GRU auto-encoder and XGBoost—with attribute-based rules expressed as policy-as-code in Open Policy Agent.

A twelve-month evaluation covering 2.1 billion production requests demonstrates that CAF increases attack-detection recall by 42 % and precision by 18 % compared with a signature Web Application Firewall and static RBAC baseline, while adding only 8 ms to the p95 gateway latency—well inside the 15 ms service-level objective required for real-time quote, billing and claim APIs. Operational metrics show a 60 % reduction in security-integration effort and a net annual benefit of $7.8 million due to prevented fraud and lower SOC triage workload. We have proven datasets, Azure ML notebooks and policy templates, demonstrating that latency-bounded, AI-augmented authorization is both technically feasible and economically compelling for enterprises pursuing Zero-Trust maturity.

Keywords: API security, Zero Trust Architecture, Contextual Authorization, Cloud-native access control, Policy-as-Code, Open Policy Agent (OPA), OAuth 2.0, OpenID Connect, Machine Learning for security, Risk-adaptive authorization, XGBoost, GRU auto encoder, Isolation Forest, Azure API Management, Real-time access control, Shared Services Platform, Financial services cyber security, AI-driven fraud prevention, Authorization latency optimization, Security integration automation

International Journal of Computer Science and Data Engineering

Hybrid Approach: A Multi-Criteria Decision-Making Method for Reverse Logistics Provider Selection

AI-Driven Risk-Adaptive Authorization for Multi-Tenant Cloud APIs - Microsoft Azure