Conference Agenda

Fossil fuels for transportation sector remain the primary contributor to global emissions, prompting an urgent exploration of renewable energy solutions, such as biodiesel. Produced from renewable feedstocks, biodiesel offers a replacement for traditional fossil fuels, helping mitigate global warming, enhance energy independence, and support rural economies. Yet, biodiesel production still faces significant challenges, including issues of energy efficiency, process optimization, and byproducts treatment, which drive up production costs and limit its broader adoption. By providing a comprehensive framework for biodiesel processes integration and waste heat and material (glycerol) valorization, this work studies to the most promising routes supporting the long-term decarbonization of the biofuels production sector. Three key feedstock, namely, refined palm oil, rapeseed oil, and soybean oil are evaluated and compared in terms of biodiesel yield. Single-step transesterification process has been modified to a two-stage transesterification approach to increase the conversion to fatty acid methyl esters under varying methanol and NaOH catalyst split fractions among two reactors. Moreover, the study addresses the efficient utilization of glycerol, a key by-product of biodiesel production, critical for improving both economic viability and environmental sustainability. To this end, several valorization routes are modeled, including crude glycerol combustion, purification to pharma-grade glycerol, supercritical water gasification, and anaerobic digestion. Mixed-integer linear programming (MILP) is employed to minimize total costs, considering both operational and capital expenditures and the constraints imposed by process integration techniques. Finally, the CO₂ emissions savings are compared to fossil fuel counterparts, including the end-use stage, demonstrating the environmental benefits of optimized biodiesel production.

Power and industrial sectors such as iron, cement, and chemical manufacturing are large-scale CO₂ emission sources, accounting for 37% of CO₂ emissions in South Korea, which should cover an 87% CO₂ reduction as of 2018. Carbon capture, utilization, and storage (CCUS) technological framework is recognized as one of the promising strategies until a perfect carbon neutral system is fully deployed. In contrast to the rapid advance of CCUS R&D (e.g., catalyst and process), the demonstration and deployment of CCUS technologies are still immature due to the limited studies on designing and planning a CCUS supply chain, especially integrating it with existing energy and industrial infrastructure. This study aims to develop a new optimization-based method to design and plan CCUS supply chain and analyze the optimal configuration and investment strategies. To achieve this goal, we develop an optimization-based approach to supply chain development using mixed-integer linear programming (MILP) model. We estimate the technical (production scale and raw material consumption), economic (unit production cost), and environmental (carbon emissions) parameters based on the literature. The objective of the optimization model is to maximize the net present value (NPV) and the net CO₂ emission (NCE) of the strategies of the CCUS supply chain under logical and practical constraints. As a real case study, Korean future CCUS system was analyzed, which includes three major CO₂ emitting industries in South Korea (power plants, steel, and chemicals) and real road transportation modes and sequestration sites. As a result, we analyzed different design and planning strategies based on various design objectives (e.g., maximizing economic and environmental benefits). In addition, by managing major cost-drivers and economic bottlenecks, we determined major decision-making problems on CCUS framework, such as sequestration vs. utilization, and provided a strategic solution of national-level planning of the CCUS supply chain. The major finding of this study can support industry stakeholders and government policymakers by providing a practical guideline to invest and plan the deployment of CCUS.

References

Suh-Young Lee, In-Beum Lee, Jeehoon Han. (2019). Design under uncertainty of carbon capture, utilization and storage infrastructure considering profit, environmental impact, and risk preference

For companies in the process and chemical industry, optimization-based scheduling is a critical advantage in today’s fast-paced and interconnected world. However, the complexity when optimizing chemical processes is high: A chemical process that requires personnel and processing equipment, consumes raw materials and utilities, and is linked to a complex supply chain, is naturally subject to many constraints and objectives [1]. Discrete Event Simulation (DES) models can describe complex real-world processes to a great level of detail, have relatively short computation times, and allow to include uncertainty parameters. However, since DES models have limited optimization capabilities, the solutions of DES models may be far away from the optimum. While MILP models enable global optimization, they quickly grow to intractable size, when trying to include all relevant constraints. In addition, MILP models can be difficult to set up, validate, and maintain for real-world applications.

The presentation shows that MILP and DES can be integrated using Benders decomposition which results in an efficient Benders-DES algorithm (BDES) that combines the strengths of rigorous optimization and high-fidelity modeling. The partial integration of DES and MILP using Benders cuts has been shown in [2,3] and is a promising line of research to combine the strengths of simulation and optimization

The developed BDES algorithm makes use of the dual information from the DES subproblem to build stronger Benders cuts. With this approach, flow shop scheduling problems with makespan minimization objective are solved, which is one of the most important operational problems in the process and chemical industry [1]. The dual information can be derived from the critical paths of the DES solutions. Critical paths play an important role in scheduling algorithms and have been used, for example, to improve the B&B procedure [4]. It is shown that the BDES algorithm is very efficient in solving random instances with 25 jobs and 5 machines, as it requires fewer DES iterations and solution time than a genetic algorithm to find near-optimal solutions.

The BDES algorithm is also effective in solving a real-word case study. The case study is a agrochemical formulation plant with seven mixing and filling lines and additional resource constraints. The BDES performs similarly to the originally proposed, monolithic-sequential MILP-DES approach [5], while requiring less modeling effort.

[1] Harjunkoski, I. et al. (2014), Computers & Chemical Engineering 62, 177

[2] Zhang et al. (2017). In: 2017 13th IEEE Conference on Automation Science and Engineering (CASE), pp. 1067–1072.

[3] Forbes et al. (2024). European Journal of Operational Research 312.3, pp. 840–854.

[4] Brucker, P. (2007). Scheduling Algorithms. 5th ed., Berlin, Germany, Springer.

[5] Wallrath, R. et al. (2023). Computers & Chemical Engineering 177, 108341.

Scheduling in the process industry determines the sequence and timing of operations to optimize objectives such as minimizing order tardiness and improving plant utilization. In research, scheduling problems are traditionally solved “batch-wise”, i.e. for an idle plant and a given set of orders, production recipes and due dates, optimal schedules are computed. However, this does not reflect reality of production planning and scheduling which is a continuous process, where new orders arrive periodically or at unknown instances, the real operations take longer or shorter periods of time than specified in the recipes, pieces or equipment break down, or operations cannot be executed as planned because resources are not available. All these aspects could be covered by infinitely fast re-computation of optimal schedules whenever an event happens or new information becomes available, but this is practically impossible for realistic problems due to the required computation time.

In online or real-time scheduling, a continuous exchange of information between the scheduling system and the control system of the production plant is necessary. The scheduling model must be updated frequently to reflect the current state of the production system and of the orders. The scheduling algorithm must react to events and disturbances fast, but also utilize the available computing power such that the schedule is near optimal.

We present an online iterative simulation-optimization approach which is tailored to handle these challenges. It builds on our previous work on simulation-optimization using evolutionary algorithms, as described in [1]. The evolutionary algorithm continuously searches for better schedules while the simulation model is updated with the latest information so that the evaluation of each generation of solutions reflects the current situation. After a pre-specified reaction time, a new solution is available after major disturbances. While the first operations of this solution are started, the schedule is further improved continuously and each assignment and timing of an operation that has not been started is based on the currently best solution.

We validate our approach using a multiproduct, multistage batch plant from the pharmaceutical industry, as in the work of Kopanos et al. [2], and demonstrate that it can generate high-quality solutions in the presence of new order arrivals and disturbances. The results are compared with those provided by an idealized clairvoyant scheduler which has access to the full information before the schedule is computed. The influence of the choice of the reaction time after a disturbance which involves a compromise between a fast reaction and better decisions in the immediate future is studied in detail.

References

[1] C. Klanke, E. Pasieka, D. Bleidorn, C. Koslwoski, C. Sonntag and S. Engell, "Evolutionary Algorithm-based Optimal Batch Production Scheduling," Computer Aided Chemical Engineering, pp. 535-540, 2022.

[2] G. M. Kopanos, C. A. Méndez and L. Puigjaner, "MIP-based decomposition strategies for large-scale scheduling problems in multiproduct multistage batch plants: A benchmark scheduling problem of the pharmaceutical industry," European Journal of Operational Research, pp. 644-655, 2010.

By 2050 around 12% of cumulative emissions reductions will come from Carbon Capture, Utilisation and Storage (CCUS) making it an essential component in the path towards net zero [1]. Focus will initially be on the retrofitting of fossil fuel power plants, which will then shift to hard-to-decarbonise industries such as iron, steel, and concrete [1]. Such industries are often grouped together in industrial clusters. Comprising both large and small point sources concentrated over a defined geographical area, industrial clusters offer an opportunity to maximise the impact of CCUS whilst also improving economic feasibility [2]. The North Sea Port (NSP) cluster is one such example of this.

Within the NSP cluster an initial set of five emitters are due to join a capture, conditioning, and transport network by 2030. From there other emitters within the area will be able to join incrementally up to 2050 [3].

However, the particular emitters who join and the timing of their connection will have a significant effect on the evolution the network. The pipeline network design will therefore have to balance design requirements for initial emitters in a backbone network, with requirements for encouraging and enabling expansion.

This study builds on scenarios defined between 2030 and 2050 [3], and applies a multi-period evolutionary-based approach (Steiner tree with Obstacles Genetic Algorithm (StObGA)) to predict pipeline year-on-year network growth in the NSP cluster. This provides a novel approach to the problem. The results are used in an examination of the potential growth of the pipeline network and an investigation of trade-offs necessary in the infrastructure design.

This work has received funding from the European Union’s Horizon 2020 Research and Innovation program under grant agreements no. 884418 (C4U project).

References:

[1] IEA, “Energy Technology Perspectives 2020,” IEA, 2020.
[2] Realise CCUS , “Industrial Clusters,” 2024. [Online]. Available: https://realiseccus.eu/ccus-and-refineries/industrial-clusters .
[3] J. O. Ejeh, S. Brown and D. Roberts, “D4.8. Report on techno-economic evaluation of the 2030 North Sea Port cluster,” C4U, 2023.

In response to growing global efforts to reduce carbon emissions, the oil refining sector—one of the largest contributors to industrial CO₂ emissions—has established ambitious decarbonization targets. Previous work [1] has developed a methodology that helps refineries choose an economically optimal decarbonization pathway. This study builds on the previous work to obtain Pareto optimal solutions for comprehensive analysis of trade-off between minimizing CO₂ emissions and minimizing costs to navigate the decarbonization journey. We use a superstructure optimization [2] framework to obtain an optimal solution, while systematically evaluating various technological pathways. A bi-criterion optimization framework is employed to generate the Pareto frontier using the epsilon-constraint method [3]. Preliminary results indicate that lower-cost, higher-emission solutions generally rely on natural gas-based technologies combined with carbon capture, while higher-cost, lower-emission solutions are linked to electric power-based technologies. Furthermore, this study incorporates a more detailed assumption regarding the carbon intensity of grid electricity, moving beyond previous assumptions of a fully decarbonized grid. By comparing decarbonization pathways under both fully decarbonized and carbon-intensive grid scenarios, we account for variations in electricity decarbonization projections based on different countries' policies and goals. This approach offers deeper insights into how the carbon profile of the grid influences optimal decarbonization strategies for refineries, with findings suggesting that carbon-intensive grids further catalyze the adoption of carbon capture technologies.

To further explore these trends, we simulate case studies across different locations, considering various projections for grid decarbonization profiles. We compare the results to assess how the Pareto frontier can inform local policy decisions and incentivize specific technologies [4] [5].

References:

[1] S. Chattopadhyay, R. Gandhi, and I. E. Grossmann, A. I. Torres "Optimization of Retrofit
Decarbonization in Oil Refineries," Foundations of Computer-Aided Process Design
(FOCAPD 2024), Breckenridge, CO, USA, Jul. 14-18, 2024.
[2] Mencarelli, L., Chen, Q., Pagot, A., & Grossmann, I. E. (2020). A review on
superstructure optimization approaches in process system engineering. Computers &
Chemical Engineering, 136, 106808. https://doi.org/10.1016/j.compchemeng.2020.106808
[3] M. Bierlaire, Optimization: Principles and Algorithms, 1st ed. Lausanne, Switzerland:
EPFL Press, 2015.
[4] Noshchenko, O., & Hagspiel, V. (2024). Environmental and economic multi-objective real
options analysis: Effective choices for field development investment planning. Energy, 135,
135037. https://doi.org/10.1016/j.energy.2024.135037
[5] Maigret, J. de, Viesi, D., Mahbub, M. S., Testi, M., Cuonzo, M., Thellufsen, J. Z.,
Østergaard, P. A., Lund, H., Baratieri, M., & Crema, L. (2022). A multi-objective optimization

Acknowledgements:

This work was financed [in part] by a grant from the Commonwealth of PA, Dept. of
Community & Economic Dev.
We acknowledge the support and funding from Shell Global and Shell Polymers Monaca