Search

• Oil product consumption and CO 2 emissions sharply influence the transport logistics indexes. • A feedback relationship is discovered for environmental pollution, indicating that oil use is not significantly driven by supply chain performance. • We found empirical sustain for the "feedback hypothesis" for Germany and Japan. This paper critically assesses the effect of fossil fuel dependence and polluting emissions from the transport sector on the performance of logistics operations in the context of Green Supply Chain Management (GSCM). We collected macro-level time-series data for a sample of 27 European Union (EU) countries over the period 2007–2018. A new Artificial Neural Networks (ANNs) algorithm is adopted in a multivariate framework to investigate the dynamic interactions among a range of Logistics Performance Indexes (LPI), the demand for oil products, and carbon dioxide (CO 2) emissions from fuel combustion in the transport sector. Empirical findings show that oil product consumption and CO 2 emissions sharply influence the transport logistics indexes. However, a feedback relationship is discovered for environmental pollution, indicating that oil use is not significantly driven by supply chain performance. Based on our empirical insights, adequate policy recommendations are provided to help turning the logistics sector towards a more sustainable path in the European area. [ABSTRACT FROM AUTHOR]

• The Belgium nuclear phase-out plan is analyzed. • A multivariate model through time-series analyses is implemented over the 1974–2019 period. • We develop and apply an innovative partial differential equations (PDE) algorithm. • Time-series and machine learning findings demonstrate suggests that the nuclear phase-out might neither adversely impact the economy's growth nor threaten the emission reduction target in the long-run. By 2025, Belgium will phase-out nuclear power. Unassessed so far, this policy reform may modify the economic and environmental channels through which energy and society interfere in this country. In this paper, we investigate whether this structural energy change may adversely impact the growth of the Belgian economy (i) and its ability to meet its long-term greenhouse gas emission targets (ii). A multivariate model comprising production factors (labor, capital, and exports), nuclear and renewable energy uses, total primary energy supply, economic growth, and CO 2 emissions from the power and heating sector is combined with real time-series data spanning the 1974–2019 period. The analysis consists in sequentially assessing two distinct nexuses (energy-economy and energy-economy-environment) over reduced- and augmented frameworks (excluding and including nuclear energy), and through a two-stage empirical strategy: time-series econometric estimations (Toda-Yamamoto causality test, Impulse Response Functions (IRFs), and the Auto-Regressive Distributed Lags (ARDL) and Machine Learning (ML) experiments with a Partial Differential Equations (PDEs) algorithm. For robustness purposes, we conduct two seminal tests which relate to dynamic predictive processes (T-Mat and Verticality tests). Besides confirming the time-series findings, our ML results highlight the necessity to timely manage the process of nuclear phase-out, along with a progressive deployment of installed renewable energy capacity. This should avoid additional economic costs, energy security threats, and undermining of climate targets. In doing so, this study combines macro-level nexus investigations with the politics and institutional determinants of nuclear energy reliance and seeks to bring inclusive knowledge on this topic. [ABSTRACT FROM AUTHOR]