In this paper, we present a comprehensive approach for evaluating efficiency in complex networks by integrating network data envelopment analysis (NDEA) with the Malmquist productivity index. The proposed method addresses the inherent challenge of accommodating negative data within the network efficiency evaluation framework, which is a common occurrence in real-world network operations. Through the introduction of a two-stage structure, the model not only effectively manages the presence of negative values, but also provides a robust and insightful assessment of network efficiency. A case study from banking sector is employed to demonstrate the efficacy of the proposed approach, showcasing its capacity to offer valuable and actionable insights for decision-making in complex network environments. The results highlight the practical applicability and importance of the extended approach in addressing the complexities associated with evaluating efficiency in diverse network settings. Manuscript profile

This paper presents a new model of two-echelon periodic supply vessel planning problem with time windows mix of facility location (PSVPTWMFL-2E) in an offshore oil and gas industry. The new mixed-integer nonlinear programming (MINLP) modelconsists ofa fleet composition problem and a location-routing problem (LRP). The aim of the model is to determine the size and type of large vessels in the first echelon and supply vessels in the second echelon.Additionally,the location of warehouse(s),optimal voyages and related schedules in both echelons are purposed.The total cost should be kept at a minimum and the need of operation regions and offshore installationsshould be fulfilled.A two-stage exact solution method, which is common for maritime transportation problems, is presented for small and medium-sized problems. In the first stage, all voyages are generated and in the second stage, optimal fleet composition, voyages and schedules are determined. Furthermore, optimal onshore base(s) to install central warehouse(s)and optimal operation region(s) to send offshore installation’s needs are decided in the second stage. Manuscript profile

Supplier selection (SS) is a decision-making process by which potential suppliers can be identified, evaluated, and ranked. Thus, multiple types of financial resources are used that can significantly contribute to the success of a firm. This study offers a broad view of SS publications from 1973 to 2019 through scientometric analysis recruiting Scopus, the Elsevier’s abstract and citation database, as a primary search engine. The documents are also statistically classified in terms of different criteria. The research results indicate that publications have considerably grown over the past few years. Moreover, the most influential countries, institutions, journals, papers, authors, and collaborations in the field of SS literature are identified. Besides, the most-cited papers are thoroughly discussed. Finally, keywords are analyzed and hot research topics are presented. This study hopes to bring awareness to researchers, journal editors, and industries in future efforts. Manuscript profile

بازده و ریسک دو عامل مهم و اساسی برای تصمیم‌گیری در حوزه مالی می‌باشند. پژوهش حاضر جهت یافتن پرتفوی بهینه برای سرمایه‌گذاری از سهام بورسی انجام گرفته و یکی‌از روش‌هایی‌که در حال حاضر محبوبیت زیادی در بین تحلیل‌گران و پژوهش-گران این حوزه شکل گرفته، روش‌های مبتنی‌بر هوش مصنوعی و در پی آن روش‌هایی با هدف کاهش سنجه‌های ریسک می‌باشد. هدف پژوهش حاضر تشکیل پرتفوی با‌استفاده از روش‌های یادگیری ماشین، سنجه ریسک و ترکیب آن با نظریه فازی است، که بازده‌ای بهتر از بازده میانگین بازار داشته باشد. خروجی هر روش وارد الگوریتم جنگل تصادفی شده و پیش‌بینی به‌وسیله این الگوریتم صورت می‌گیرد و در مرحله آخر، خروجی پیش‌بینی‌ برای تشکیل سبد سرمایه وارد مدل بهینه‌سازی ارزش در معرض ریسک و ارزش در معرض ریسک شرطی با رویکرد نظریه فازی می‌شوند. اطلاعات سهم‌ها به‌صورت روزانه و بازه زمانی آن از ابتدای سال 1394 تا اواسط سال 1398 می‌باشد. در پایان هرکدام از این روش‌ها و مراحل با بازده واقعی بازار مقایسه گردید. بر اساس نتایج بدست آمده سنجه‌ریسک CVAR قابلیت بهتری را نسبت‌به سنجه ریسک VAR داشته است، هم‌چنین الگوریتم جنگل تصادفی در بین الگوریتم‌های یادگیری ماشین استفاده شده، نتایج بهتری را در انتخاب سبد سرمایه‌گذاری رقم زده‌ است. Manuscript profile