•An introduction to optimization under uncertainty is presented.•Recent advances in data-driven optimization under uncertainty are reviewed.•Future perspective on a “closed-loop” data-driven ...optimization framework is given.•Potential application of deep learning to data-driven optimization is discussed.
This paper reviews recent advances in the field of optimization under uncertainty via a modern data lens, highlights key research challenges and promise of data-driven optimization that organically integrates machine learning and mathematical programming for decision-making under uncertainty, and identifies potential research opportunities. A brief review of classical mathematical programming techniques for hedging against uncertainty is first presented, along with their wide spectrum of applications in Process Systems Engineering. A comprehensive review and classification of the relevant publications on data-driven distributionally robust optimization, data-driven chance constrained program, data-driven robust optimization, and data-driven scenario-based optimization is then presented. This paper also identifies fertile avenues for future research that focuses on a closed-loop data-driven optimization framework, which allows the feedback from mathematical programming to machine learning, as well as scenario-based optimization leveraging the power of deep learning techniques. Perspectives on online learning-based data-driven multistage optimization with a learning-while-optimizing scheme are presented.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
This paper addresses the optimal design and planning of biomass-to-liquids (BTL) supply chains under economic and environmental criteria. The supply chain consists of multisite ...distributed–centralized processing networks for biomass conversion and liquid transportation fuel production. The economic objective is measured by the total annualized cost, and the measure of environmental performance is the life cycle greenhouse gas emissions. A multiobjective, multiperiod, mixed-integer linear programming model is proposed that takes into account diverse conversion pathways and technologies, feedstock seasonality, geographical diversity, biomass degradation, infrastructure compatibility, demand distribution, and government incentives. The model simultaneously predicts the optimal network design, facility location, technology selection, capital investment, production planning, inventory control, and logistics management decisions. The problem is formulated as a bicriterion optimization model and solved with the ε-constraint method. The resulting Pareto-optimal curve reveals how the optimal annualized cost and the BTL processing network structure change with different environmental performances of the supply chain. The proposed approach is illustrated through a county-level case study for the state of Iowa.
Full text
Available for:
IJS, KILJ, NUK, PNG, UL, UM
Since 2020, the COVID-19 pandemic has urged event holders to shift conferences online. Virtual and hybrid conferences are greener alternatives to in-person conferences, yet their environmental ...sustainability has not been fully assessed. Considering food, accommodation, preparation, execution, information and communication technology, and transportation, here we report comparative life cycle assessment results of in-person, virtual, and hybrid conferences and consider carbon footprint trade-offs between in-person participation and hybrid conferences. We find that transitioning from in-person to virtual conferencing can substantially reduce the carbon footprint by 94% and energy use by 90%. For the sake of maintaining more than 50% of in-person participation, carefully selected hubs for hybrid conferences have the potential to slash carbon footprint and energy use by two-thirds. Furthermore, switching the dietary type of future conferences to plant-based diets and improving energy efficiencies of the information and communication technology sector can further reduce the carbon footprint of virtual conferences.
In this work, we perform a comparative techno-economic and environmental analysis for manufacturing ethylene and propylene from naphtha and from shale gas with rich natural gas liquids (NGLs). We ...first propose two novel process designs for producing ethylene and propylene from NGLs-rich shale gas. These two designs employ steam co-cracking of an ethane–propane mixture and an integration of ethane steam cracking and propane dehydrogenation, respectively. For benchmarking, we also consider a conventional process design in which ethylene and propylene are produced via steam cracking of naphtha. Detailed process models are developed for all the three designs to obtain the mass and energy balances of each unit operation. On this basis, techno-economic analysis and life cycle analysis are conducted for each of the three designs in order to systematically compare the production costs and life cycle environmental impacts of ethylene and propylene manufactured from shale gas and naphtha based on the same conditions. The economic analysis indicates that manufacturing ethylene and propylene from NGLs-rich shale gas is more attractive than from naphtha. The environmental impact analysis shows that manufacturing ethylene and propylene from NGLs-rich shale gas results in lower life cycle water consumption but higher life cycle greenhouse gas emissions.
Full text
Available for:
IJS, KILJ, NUK, PNG, UL, UM
•Novel quantum computing based hybrid solution strategies are developed.•The proposed hybrid techniques leverage both quantum and classical computers.•Applications ranging from molecular design to ...logistics optimization are addressed.•Application problems across multiple scales are solved by respective hybrid method.•Hybrid techniques outperform general-purpose state-of-the-art deterministic solvers.
Quantum computing (QC) has gained popularity due to its unique capabilities that are quite different from that of classical computers in terms of speed and methods of operations. This paper proposes hybrid models and methods that effectively leverage the complementary strengths of deterministic algorithms and QC techniques to overcome combinatorial complexity for solving large-scale mixed-integer programming problems. Four applications, namely the molecular conformation problem, job-shop scheduling problem, manufacturing cell formation problem, and the vehicle routing problem, are specifically addressed. Large-scale instances of these application problems across multiple scales ranging from molecular design to logistics optimization are computationally challenging for deterministic optimization algorithms on classical computers. To address the computational challenges, hybrid QC-based algorithms are proposed and extensive computational experimental results are presented to demonstrate their applicability and efficiency. The proposed QC-based solution strategies enjoy high computational efficiency in terms of solution quality and computation time, by utilizing the unique features of both classical and quantum computers.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Safe, efficient, and sustainable operations and control are primary objectives in industrial manufacturing processes. State-of-the-art technologies heavily rely on human intervention, thereby showing ...apparent limitations in practice. The burgeoning era of big data is influencing the process industries tremendously, providing unprecedented opportunities to achieve smart manufacturing. This kind of manufacturing requires machines to not only be capable of relieving humans from intensive physical work, but also be effective in taking on intellectual labor and even producing innovations on their own. To attain this goal, data analytics and machine learning are indispensable. In this paper, we review recent advances in data analytics and machine learning applied to the monitoring, control, and optimization of industrial processes, paying particular attention to the interpretability and functionality of machine learning models. By analyzing the gap between practical requirements and the current research status, promising future research directions are identified.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
•Bilevel MINLP model for design and planning of non-cooperative supply chains.•Stackelberg game and generalized Nash equilibrium in multi-echelon supply chains.•Global optimization by KKT ...transformation and improved branch-and-refine algorithm.•County-level case study on a potential biofuel supply chain in Illinois.
We propose a bilevel mixed-integer nonlinear programming (MINLP) model for the optimal design and planning of non-cooperative supply chains from the manufacturer's perspective. Interactions among the supply chain participants are captured through a single-leader–multiple-follower Stackelberg game under the generalized Nash equilibrium assumption. Given a three-echelon superstructure, the lead manufacturer in the middle echelon first optimizes its design and operational decisions, including facility location, sizing, and technology selection, material input/output and price setting. The following suppliers and customers in the upstream and downstream then optimize their transactions with the manufacturer to maximize their individual profits. By replacing the lower level linear programs with their KKT conditions, we transform the bilevel MINLP into a single-level nonconvex MINLP, which is further globally optimized using an improved branch-and-refine algorithm. We also present two case studies, including a county-level biofuel supply chain in Illinois, to illustrate the application of the proposed modeling and solution methods.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK
The globalized supply chain for crystalline silicon (c-Si) photovoltaic (PV) panels is increasingly fragile, as the now-mundane freight crisis and other geopolitical risks threaten to postpone major ...PV projects. Here, we study and report the results of climate change implications of reshoring solar panel manufacturing as a robust and resilient strategy to reduce reliance on foreign PV panel supplies. We project that if the U.S. could fully bring c-Si PV panel manufacturing back home by 2035, the estimated greenhouse gas emissions and energy consumption would be 30% and 13% lower, respectively, than having relied on global imports in 2020, as solar power emerges as a major renewable energy source. If the reshored manufacturing target is achieved by 2050, the climate change and energy impacts would be further reduced by 33% and 17%, compared to the 2020 level. The reshored manufacturing demonstrates significant progress in domestic competitiveness and toward decarbonization goals, and the positive reductions in climate change impacts align with the climate target.
The past few years have witnessed a rapid evolution of perovskite solar cells, an unprecedented photovoltaic (PV) technology with both relatively low cost and high power conversion efficiency. In ...this paper, we perform a life cycle assessment for two types of solution-processed perovskite solar modules to shed light on the environmental performance of this promising class of PVs. One module is equipped with FTO glass, a gold cathode, and mesoporous TiO sub(2) scaffold; the other is equipped with ITO glass, a silver cathode, and ZnO thin film. We develop comprehensive life cycle inventories (LCIs) for all components used in the modules. Based on the LCI results, we conduct life cycle impact assessment for 16 common life cycle impact indicators, Eco-indicator 99, and two sustainable indicators: the energy payback time (EPBT) and the CO sub(2) emission factor. We compare the results of Eco-indicator 99, the EPBT, and the CO sub(2) emission factor among existing PV technologies, and further perform uncertainty analysis and sensitivity analysis for the two modules. The results demonstrate that perovskite solar modules possess the shortest EPBT, and future research should be directed to improving the system performance ratio and the device lifetime, and reducing precious metal consumption and energy-intensive operations in order to lower the CO sub(2) emission factor.
Display omitted
•Recent developments in deep learning for inverse molecular design are reviewed.•Challenges and opportunities for three directions are identified and discussed.•GNNs are seen as ...transformative catalyst towards developing 3D representations.•Chemical text mining and QC computations can contribute to growing property data.•Progress in generative models and RL is poised to improve current design methods.
The discovery of superior molecular solutions through computational methods is critical for innovative technologies and their role in addressing pressing resources, health, and environmental issues. Despite its short timespan, the synergetic application of deep learning to inverse molecular design has outpaced decades of theoretical efforts, bearing promise to transform current molecular design paradigms. Herein, we provide an overview of the element of computational inverse molecular design and offer our views on current limitations and outstanding challenges. In our perspective, three main directions are identified for each element and analyzed in terms of their merits and relevant novel deep learning developments. For the molecular representations element, Graph Neural Networks (GNNs), grids, and knowledge graphs (KGs) are discussed for enhancing the expressivity, complexity, descriptivity of relevant molecular information, respectively. Second, chemical text mining, accelerated quantum chemical calculations, and transfer learning are explored to augment the size and the accuracy of current property data and predictive models. Last, emerging trends in design methods including generative modeling, reinforcement learning (RL), and active learning (AL) are examined for optimizing not only computational costs, but also experimental and simulation efforts. The presented discussions are aimed at catalyzing progress and interdisciplinary collaborations toward general-purpose inverse design frameworks.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP