DOI: 10.3303/CET25120014 Paper Received: 15 May 2025; Revised: 15 September 2025; Accepted: 3 November 2025 Please cite this article as: Chua L.H., San Juan J.L., 2025, Optimization of Dynamically Adjusting Segregation Strategies in Wastewater Treatment Plants, Chemical Engineering Transactions, 120, 79-84 DOI:10.3303/CET25120014 CHEMICAL ENGINEERING TRANSACTIONS VOL. 120, 2025 A publication of The Italian Association of Chemical Engineering Online at www.cetjournal.it Guest Editors: Bing Shen How, Viknesh Andiappan, Denny K.S. Ng, Hon Loong Lam, Petar S. Varbanov Copyright © 2025, AIDIC Servizi S.r.l. ISBN 979-12-81206-21-2; ISSN 2283-9216 Optimization of Dynamically Adjusting Segregation Strategies in Wastewater Treatment Plants Lance Harley Chua, Jayne Lois San Juan* Department of Industrial and Systems Engineering, De La Salle University, 1004 Manila, Philippines bAffiliation and jayne.sanjuan@dlsu.edu.ph Wastewater treatment is critical for preserving the limited global supply of clean water but remains highly energy- intensive, with growing demand driving increased fossil fuel consumption in WWTPs. This study examines the integration of a dynamic segregation system that adjusts in real-time to fluctuations in water quality and flow. By minimizing unnecessary treatments, the system improves efficiency and reduces energy use. Literature shows that water quality variability is common, and periodic decision-making adjustments enhance treatment effectiveness. As regulatory penalties and inefficiencies grow more costly, dynamic segregation presents a viable long-term solution for sustainable and cost-effective wastewater management. 1. Introduction Only 0.5 % of Earth’s water is readily usable by humans (Baker et al., 2016). Contamination from urban, agricultural, and industrial activities has further strained this limited supply, making wastewater treatment essential for environmental protection and water security. Conventional treatment processes, however, are highly energy intensive. In 2015, China’s activated sludge systems alone consumed over 25.15 B kWh (Zhang et al., 2020). More broadly, wastewater treatment accounts for about 35 % of municipal energy consumption (Masłoń, 2017) and contributes over 50 % of greenhouse gas emissions within the water sector (Nakkasunchi, 2020). These figures emphasize the urgency of pursuing more energy-efficient treatment strategies. Wastewater segregation has emerged as a promising approach, reducing energy use by routing influent based on contaminant profiles and avoiding overtreatment (Ranade et al., 2014). Other energy recovery options, such as microalgae cultivation, have also been explored, although their effectiveness depends on environmental variables like natural light (Singh et al., 2024). Water quality can also fluctuate due to seasonal changes or the presence of emerging contaminants like antibiotics and algae (Caligan et al., 2021), affecting removal efficiency and treatment reliability (Guo et al., 2025). These challenges are particularly pronounced with the rise of decentralized treatment systems (Leigh and Lee, 2019), where variability has a larger operational impact (Van De Walle et al., 2022). Real-time monitoring tools, such as IoT sensors, can provide continuous influent quality data for operational decision-making (Goncalves et al., 2020). For instance, AI-based control systems have been shown to improve treatment efficiency by adjusting oxidant dosing in response to changing influent conditions (Pham et al., 2022). Liew et al. (2014) extended water pinch to multiperiod planning but retained a fixed network, used a single aggregated water quality parameter, and focused solely on freshwater minimization. Other works, such as Caligan et al. (2022), also prioritize water use minimization without explicitly modeling operational trade-offs between different segregation strategies. This study addresses these gaps by developing a multiperiod model tailored to segregation-based WWTPs that integrates treatment-stream-specific efficiencies, energy use, freshwater intake, and segregation costs into a unified cost-minimization framework. Unlike multiperiod water pinch, the proposed method allows routing to change each period in response to contaminant profiles and cost priorities and directly compares fixed versus dynamic segregation within the same system to quantify the operational and economic benefits of flexibility. 79 2. System Definition The optimization model incorporates operational, energy, and environmental costs, including freshwater usage and release. Building on Sa’ad et al. (2022) and Alfaisal (2024), it enables dynamic segregation of wastewater from multiple sources based on time-varying flows and contaminant levels, routing them to treatment streams with different removal efficiencies. An aggregate quality level is assumed for the treated water, serving as the basis for performance comparisons (San Juan et al., 2020). Freshwater is added as needed to meet volume and dilution requirements. The model adapts allocations over time to respond to quality fluctuations, subject to contaminant limits, treatment capacities, and mass balance constraints. 3. Model Formulation 3.1 Assumptions In the dynamic segregation system, daily water flow and quality are segmented across multiple time periods rather than averaged over the entire day. All incoming wastewater is assumed to undergo treatment, and any treated volume exceeding the demand is discarded from the system. Freshwater, when introduced, is assumed to be free of contaminants. 3.2 Objective Function and Cost Components The objective function in Eq(1) minimizes total treatment and water handling costs, including the cost of introducing and discharging water. Energy costs in Eq(2) depend on the segregation strategy and treatment chosen. Introduction and discharge costs are computed per liter, as shown in Eq(3) and Eq(4). min Z = EC + WC + XC (1) EC = (EDyn · X + Σt ETrt) · EP (2) WC = Σt FWt · WP (3) XC = Σt XWt · XP (4) 3.3 Water Segregation Wastewater from each source is routed to specific treatment streams. Eq(5) ensures source allocations match total inflow. Eq(6) and Eq(7) compute total water and contaminant mass per stream, while Eq(8) sets a limit on stream capacity. If fixed allocation is selected, Eq(9) and Eq(10) lock percentage distributions to prevent variation over time. Σp Qipt = 1, ∀ i, t (5) Wpt = Σi Qipt Wit, ∀ p, t (6) Cpkt = Σi Qipt Cikt, ∀ p, k, t (7) Wpt + Σk Cpkt ≤ Wmaxp, ∀ p, t (8) Qipt ≥ Qfixip - X, ∀ i, p, t (9) Qipt ≤ Qfixip + X, ∀ i, p, t (10) 3.4 Water Treatment Wastewater that enters treatment undergoes a process where a given percentage of contaminant content is removed from a given stream, based on the removal efficiency, given in Eq(11). Eq(12) computes the energy consumption at any given period by multiplying the energy consumption with unit flow for each stream, summing them to get the total energy consumption. Coutpkt = Cpkt (1 - Rpk), ∀ p, k, t (11) ETrt = Σp ((Wpt + Σk Cpkt) · EConsp) (12) 80 3.5 Water Distribution Treated water is then aggregated, summing up the water and contaminant content from all streams, alongside introduced freshwater. Safety standards for contaminants are depicted in Eq(15), while tank capacity is listed in Eq(16). Lastly, water that is not used to satisfy demand is then released. WDt = Σp Wpt + FWt, ∀ t (13) CDkt = Σp Cpkt, ∀ k, t (14) CDkt ≤ ConcLimitk · (WDt + Σk CDkt), ∀ k, t (15) WDt + Σk CDkt ≤ CAP, ∀ t (16) XWt = WDt + Σk CDkt - Demt, ∀ t (17) 4. Illustrative Case Study This case study examines three wastewater sources, each containing varying amounts of three contaminant types that fluctuate over 24 periods representing the hours in a day. The model can be adapted for other practical scenarios by adjusting the period length, for example, using 30 periods to represent daily variations over a month, or subdividing each hour into finer intervals for more dynamic, high-resolution adjustments. Figure 1 illustrates the general level of water flow from each source. For each source, one contaminant was selected to have a higher mass flow rate, while the other contaminants are to have a similar lower mass flow rate. Likewise, three treatment streams (collectors) are available, each with varying removal efficiencies across the three contaminant types. The cost to treat 1 Liter of wastewater also differs per stream, and these values are summarized in Table 1. Figure 2 shows that freshwater is added only when needed to meet volume requirements, due to its associated cost. Treated water consistently meets quality standards, indicating that dilution is unnecessary. This confirms that treatment streams alone are sufficient for contaminant removal, and freshwater is introduced solely to meet demand. Figure 3 shows the allocation of wastewater from each source to different treatment streams over time. Two key factors drive these shifts: fluctuations in source water flow and changes in demand. During periods of lower demand, sources are primarily routed to the streams that are most effective at removing their dominant contaminants. As demand increases and freshwater supplementation becomes necessary, allocation shifts toward streams with the lowest treatment cost per L, prioritizing volume over efficiency. When demand tapers off, the system gradually returns to prioritizing contaminant-specific removal efficiency. Table 1: Removal Efficiency and Cost per L of Treatment Streams Efficiency Contaminant 1 Contaminant 2 Contaminant 3 Cost per L Stream 1 0.95 0.70 0.70 0.034 Stream 2 0.70 0.95 0.70 0.017 Stream 3 0.70 0.70 0.95 0.0255 Figure 1: Water Flow in L from Sources 1 to 3 over 24 periods 0 100 200 300 400 500 600 700 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 M a s s F lo w ( L ) Period Source 1 Source 2 Source 3 81 Figure 2: Water Intake in L per period Figure 3: Allocation of Wastewater from Sources to Treatment Streams in L The proposed dynamic system yields an energy cost of $ 899.94, freshwater costs of $ 8,622.80, and a water release cost of $ 48.08, totalling $ 9,570.82. This will be utilized as the basis for comparing other scenarios or decisions that would affect the system. 5. Scenario Analysis 5.1 Forced Fixed Segregation Standard optimization models, such as Saad et al. (2022), typically assume fixed segregation infrastructure. In contrast, this study compares fixed and dynamic systems, with cost differences detailed in Table 2, including the percentage increase associated with fixed segregation. The results highlight the potential of dynamic systems to achieve meaningful operational cost reductions as seen in small optimizations that can have a significant impact given the high cost of developing and maintaining WWTPs. Table 2: Cost Comparison between Dynamic and Fixed Segregation Systems Scenario Energy Cost Freshwater Cost Water Release Cost Total Cost Dynamic $ 899.94 $ 8,622.80 $ 48.08 $ 9,570.82 Fixed $ 911.28 $ 8,629.34 $ 48.36 $ 9,588.98 % Increase 1.26 % 0.07 % 0.58 % 0.02 % 5.2 Sustainability and Reduced Water Usage To conserve limited water resources, reuse strategies are being explored. Jivani et al. (2025) propose using slightly contaminated water for industrial processes to reduce freshwater demand. However, with contaminant discharge unchanged, reduced flow (20 % lower in all periods) leads to higher concentrations. As shown in Table 3, this raises freshwater intake costs but lowers energy and treatment costs. While this approach offers potential long-term savings, especially since energy can potentially make up around 40 % of WWTP costs (Saghafi et al., 2016), it may increase treatment difficulty due to more concentrated waste. 0 1,000 2,000 3,000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Li te rs Period Inflow Demand Intake Contaminant 1 Contaminant 2 Contaminant 3 Total 0 200 400 600 800 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 A llo c a ti o n i n L Period Source 1 - Stream 1 Source 1 - Stream 2 Source 1 - Stream 3 Source 2 - Stream 1 Source 2 - Stream 2 Source 2 - Stream 3 82 As shown in Figure 4, rising contaminant concentrations require the addition of freshwater to dilute the effluent to acceptable levels. This results in a total fluid volume that exceeds actual demand, illustrating a trade-off between contamination control and resource efficiency. In such cases, introducing freshwater is often more cost-effective than applying stronger, more expensive treatment methods (Nemet et al., 2021). Table 3: Cost Comparison between Original and Reduced Water Flow Water Flow Energy Cost Freshwater Cost Water Release Cost Total Cost Original $ 899.94 $ 8,622.80 $ 48.08 $ 9,570.82 20 % Reduction $ 664.30 $ 13,252.80 $ 22.63 $ 13,939.73 Figure 4: Freshwater intake patterns for reduced water flow rate, in L per period Comparing the costs to the use of a fixed segregation system as seen in Table 4, a 1 % increase is observed for energy and water release costs, while the costs of introducing freshwater remain unchanged. Table 4: Cost Comparison between Dynamic and Fixed Segregation Systems Scenario Energy Cost Freshwater Cost Water Release Cost Total Cost Dynamic $ 664.30 $ 13,252.80 $ 22.63 $ 13,939.73 Fixed $ 673.37 $ 13,252.80 $ 22.90 $ 13,949.06 % Increase 1.37 % 0.00 % 1.19 % 0.07 % 6. Conclusions This study presents an optimization model for wastewater treatment plants that incorporates IoT-enabled dynamic segregation to enhance blue energy recovery. Unlike fixed systems, dynamic segregation offers greater flexibility and cost efficiency by adapting to fluctuations in wastewater quality and flow. By directing wastewater to the most suitable treatment streams, the system reduces energy use and improves contaminant removal, especially valuable in settings with high pollution levels or strict environmental regulations. Although IoT integration requires higher upfront investment, long-term savings make it economically viable for industrial and agricultural zones. Future research could address limitations related to contaminant interactions, which may hinder treatment or cause adverse chemical reactions. Exploring the integration of multiple time-sensitive energy recovery methods such as algae biofuel and employing advanced tools such as nonlinear programming or system dynamics modeling may further improve system performance under variable conditions. Nomenclature Z – total cost, $ EC – total energy costs, $ WC– total freshwater costs, $ XC– total water release costs, $ EP – cost per unit energy, $/kWh WP – cost per unit of freshwater, $/L XP – cost per released freshwater, $/L EDyn – energy used in dynamic segregation, kWh ETrt – treatment energy per period, $ X – utilization of fixed or dynamic segregation, binary FWt – introduced freshwater, L XWt – released freshwater, L Qipt – % allocation between sources and streams, - Wpt – water content before treatment, L Cpkt – contaminant content before treatment, L Wmaxp – maximum flow rate per period, L 0 500 1,000 1,500 2,000 2,500 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 L it e rs Period Inflow Demand Intake Contaminant 1 Contaminant 2 Contaminant 3 Total 83 Qfixip – fixed value for allocation, - Coutpkt – contaminant content after treatment, L Rpk – contaminant removal efficiency, - EConsp – energy consumption in treatment, kWh/L WDt – water content in distribution pool, L CDkt – contaminant content in distribution pool, L ConcLimitk – maximum concentration of contaminant, - Demt – water demand, L References Alfaisal F.M., 2024, Development of a sustainable optimization model for planning regional wastewater systems with consideration of water quality. AIP Advances, 14, 105303. Baker B., Aldridge C., Omer A., 2016, Water: Availability and use. Mississippi State University Extension. 3011. accessed 14.08.2025. Caligan C.J.A., Garcia M.M.S., Mitra J.L., Mayol A.P., San Juan J.L.G., Culaba A.B., 2020, Multi-objective optimization of water exchanges between a wastewater treatment facility and algal biofuel production plant. IOP Conference Series: Earth and Environmental Science, 463, 012050. Caligan C.J.A., Garcia M.M.S., Mitra J.L., Mayol A.P., San Juan J.L.G, 2022, Multi-objective optimization for a wastewater treatment plant and sludge-to-energy network. Journal of Cleaner Production, 368, 133047. Damalerio R.G., Beltran A.B., Orbecido A.H., Aviso K.B., Tanhueco R.M.T., Torneo A.R., Ancog R.C., Promentilla M.A.B., 2022, Decentralised wastewater management systems in developing countries: Key barriers and potential resource recovery applications. Chemical Engineering Transactions, 97, 367–372. Gonçalves R., Soares J.J.M., Lima R.M.F., 2020, An IoT-based framework for smart water supply systems management. Future Internet, 12, 114. Guo X., Lu X., Jia J., Xing D., Li Y., Cao G., Zhang Z., 2025, Comprehensive assessment of 45 antibiotics in ten urban wastewater treatment plants in Northeastern China: Terminal treatment is not a reliable guard. Journal of Hazardous Materials, 489, 137755. Jivani J., Srinivasarao M., Dhanwani A., 2025, Modified sustainable water management strategy in batch process plants: A case study on reducing freshwater consumption. Water Supply, 25, 212–227. Leigh N.G., Lee H., 2019, Sustainable and resilient urban water systems: The role of decentralization and planning. Sustainability, 11, 918. Liew P.Y., Wan Alwi S.R., Manan Z.A., 2014, Multiperiod planning of water networks involving batch processes, Chemical Engineering Transactions, 39, 1573–1578. Masłoń A., 2017, Analysis of energy consumption at the Rzeszów Wastewater Treatment Plant. E3S Web of Conferences, 22, 00115. Nakkasunchi S., Hewitt N.J., Zoppi C., Brandoni C., 2020, A review of energy optimization modelling tools for the decarbonisation of wastewater treatment plants. Journal of Cleaner Production, 279, 123811. Nemet A., Bogataj M., Kravanja Z., 2021, Optimization of the sequence of wastewater treatment in the cosmetic industry. Chemical Engineering Transactions, 88, 715–720. Pham C.M., Pham N.Q., Le A.K., 2022, Oxidation-reduction potential and peroxone process in antibiotic residues removal from hospital wastewater. Chemical Engineering Transactions, 97, 187–192. Ranade V.V., Bhandari V.M., 2014, Industrial wastewater treatment, recycling, and reuse. Butterworth- Heinemann/Elsevier, Pune, India. , accessed 15.08.2025. Sa’ad S.F., Alwi S.R.W., Lim J.S., Manan Z.A., 2022, The economic study of centralised water reuse exchange system in the industrial park considering wastewater segregation. Computers and Chemical Engineering, 164, 107863. Saghafi S., Mehrdadi N., Bid Hendy G.N., Rad H.A., 2016, Estimating the electrical energy in different processes for Nasir Abad Industrial Wastewater Treatment Plant with emphasis on COD removal. Journal of Environmental Studies, 42, 4-6. San Juan J.L., Caligan C.J., Garcia M.M., Mitra J., Mayol A.P., Sy C., Ubando A., Culaba A., 2020, Multi- Objective Optimization of an Integrated Algal and Sludge-Based Bioenergy Park and Wastewater Treatment System. Sustainability, 12, 7793. Singh N., Rathilal S., 2024, Cultivation of Scenedesmus sp., Chlorella vulgaris and Spirulina platensis in sewage industrial wastewater: Bench-scale optimization for biofuel production. Chemical Engineering Transactions, 113, 409–414. Van De Walle A., Torfs E., Gaublomme D., Rabaey K., 2022, In silico assessment of household level closed water cycles: Towards extreme decentralization. Environmental Science and Ecotechnology, 10, 100148. 84 http://journal/ 0049.pdf Optimization of Dynamically Adjusting Segregation Strategies in Wastewater Treatment Plants