Industrial-environmental management in China's iron and steel industry under multiple objectives and uncertainties

Abstract

Industrial-environmental management is a multi-objective optimization problem plagued with multiple uncertainties. Most studies only optimize few objectives and often neglect these uncertainties. This study builds a 6-objective optimization problem to quantify energy conservation and emission reduction (ECER) potentials in China's iron and steel industry. First, uncertainties are simulated through 100,000-time random sampling, NSGA-II and the mean-effective objective mechanism are applied to calculate optimal solutions. Finally, a global sensitivity analysis is performed to classify uncertainty parameters based on their impacts on objectives' performance. Results show: (1) There exist significant discrepancies between objectives' performance under certainty and uncertainty. For example, the deterministic CO2 intensity is 1148 kg/t, which is 11.93% lower than its value under uncertainty. Therefore, neglecting uncertainty increases the risk of noncompliance with policy targets as they might be too strict; (2) Two critical uncertainty parameters (steel ratios and technology penetration rates) have the most severe impacts on objectives' performance, hence, reducing their fluctuation can minimize uncertainties when estimating ECER potentials; (3) By-product recycling and energy efficiency measures have good performance in all objectives, thus, should be prioritized. Moreover, from 77-strategies assessed, 11 are identified as key-strategies due to their large ECER effects, hence, should be strongly promoted.