Drinking water distribution systems are increasingly vulnerable to sewage intrusion due to aging water infrastructure and intensifying water stress. While the health risks associated with sewage intrusion have been extensively studied, little is known about the impacts of intruded bacteria and dissolved organic matter (DOM) on microbiology in drinking water. In this dynamic study, we demonstrate that the intrusion of 1 % sewage into tap water resulted in immediate contamination, including an 8-fold increase in biomass (TCC), a 48.9 % increase in bacterial species (ASVs), a 12.5 % increase in organic carbon content (DOC), and a 13.5 % increase in unique DOM molecular formulae. Over time, sewage intrusion altered tap water microbiology by accelerating bacterial growth rates (5-fold faster), selectively promoting ASVs in community succession, and producing 998 more unique DOM formulae. More significantly, statistical analysis revealed that the intrusion of 1 % sewage shifted the driving force of bacterial and DOM composition covariance from a DOM-dependent process in tap water to a bacterial-governed process post-intrusion. Our results clearly demonstrate the disruptive effects of sewage intrusion into tap water, emphasizing the urgent need to consider the long-lasting impacts of sewage intrusion in drinking water distribution systems, in addition to its immediate health risks.

@en

Although simulated studies have provided valuable knowledge regarding the communities of planktonic bacteria and biofilms, the lack of systematic field studies have hampered the understanding of microbiology in real-world service lines and premise plumbing. In this study, the bacterial communities of water and biofilm were explored, with a special focus on the lifetime development of biofilm communities and their key influencing factors. The 16S rRNA gene sequencing results showed that both the planktonic bacteria and biofilm were dominated by Proteobacteria. Among the 15,084 observed amplicon sequence variants (ASVs), the 33 core ASVs covered 72.8 %, while the 12 shared core ASVs accounted for 62.2 % of the total sequences. Remarkably, it was found that the species richness and diversity of biofilm communities correlated with pipe age. The relative abundance of ASV2 (f_Sphingomonadaceae) was lower for pipe ages 40–50 years (7.9 %) than for pipe ages 10–20 years (59.3 %), while the relative abundance of ASV10 (f_Hyphomonadaceae) was higher for pipe ages 40–50 years (19.5 %) than its presence at pipe ages 20–30 years (1.9 %). The community of the premise plumbing biofilm had significantly higher species richness and diversity than that of the service line, while the steel-plastics composite pipe interior lined with polyethylene (S-PE) harbored significantly more diverse biofilm than the galvanized steel pipes (S-Zn). Interestingly, S-PE was enriched with ASV27 (g_Mycobacterium), while S-Zn pipes were enriched with ASV13 (g_Pseudomonas). Moreover, the network analysis showed that five rare ASVs, not core ASVs, were keystone members in biofilm communities, indicating the importance of rare members in the function and stability of biofilm communities. This manuscript provides novel insights into real-world service lines and premise plumbing microbiology, regarding lifetime dynamics (pipe age 10–50 years), and the influences of pipe types (premise plumbing vs. service line) and pipe materials (S-Zn vs. S-PE).

@en

Treated drinking water is delivered to customers through drinking water distribution systems (DWDSs). Although studies have focused on exploring the microbial ecology of DWDSs, knowledge about the effects of different water treatments on the bacterial community of biofilm and loose deposits in DWDS is limited. This study assessed the effects of additional treatments on the bacterial communities developed in 10 months’ old pilot DWDSs. The results showed a similar bacterial community in the pipe-wall biofilm, which was dominated by Novosphingobium spp. (20–82 %) and Sphingomonas spp. (11–53 %), regardless of the treatment applied. The bacterial communities that were retained in the distribution systems (including pipe-wall biofilm and loose deposits) were similar to the particle-associated bacteria (PAB) in the corresponding supply water. The additional treatments showed clear effects of the removal and/or introduction of particles. The genera Aeromonas spp., Clostridium spp., Legionella spp., and Pseudomonas spp., which contain opportunistic pathogenic species, were only detected among the PAB in ion exchange system. Our study demonstrated that the biofilm community is consistent across treatments, and the contribution from bacteria in loose deposits is important but can be controlled by removing particles. These findings offer more insight into the origin and development of microbial ecology in DWDSs and suggest paths for further research on the possibility of managing the microbial ecology in distribution systems.

@en

Pipe materials appear to play an important role in the development of biofilms in drinking water distribution systems. However, there is controversy as to whether pipe materials shape the composition and diversity of bacterial communities in biofilms. To investigate the long-term effects of pipe materials on biofilms, triplicate samples of mature biofilms on unplasticized polyvinyl chloride (PVC-U), grey cast iron and asbestos cement (pipe age > 40 years) were obtained from three areas of an unchlorinated drinking water distribution system in the Netherlands. Illumina sequencing was performed and 773 OTUs (730 OTUs-814 OTUs) were detected within the biofilms on the three pipe materials, all of which were dominated by Proteobacteria (36.2%-46.1%). Both the alpha and beta diversity results showed that the bacterial communities of the biofilms formed on different pipe materials were highly similar. The neutral community model revealed that the assembly of the biofilm communities was governed by environmental selection rather than neutral processes. Among the 142 shared OTUs between the water and biofilm samples, there were 25 enriched OTUs (e.g., OTU7, assigned as Nitrospira spp.), which accounted for 62.6% of the total sequences, while 16 OTUs were disadvantaged (e.g., OTU14 and OTU40, assigned as Hyphomicrobiaceae), accounting for 2.2% of the sequences. Based on the findings, we propose and discuss a harmonisation process by which biofilms with significant differences due to the pipe material harmonize over time resulting in biofilms with similar bacterial communities. Our findings provide valuable insights into long-term biofilm development, bridging an essential gap in our current understanding of the influence of pipe materials on biofilm communities. These findings also highlight the importance of long-term studies and point to a potentially masked harmonizing process during biofilm development over years/decades.

@en

Premise plumbing plays an essential role in determining the final quality of drinking water consumed by customers. However, little is known about the influences of plumbing configuration on water quality changes. This study selected parallel premise plumbing in the same building with different configurations, i.e., laboratory and toilet plumbing. Water quality deteriorations induced by premise plumbing under regular and interrupted water supply were investigated. The results showed that most of the water quality parameters did not vary under regular supply, except Zn, which was significantly increased by laboratory plumbing (78.2 to 260.7 µg/l). For the bacterial community, the Chao1 index was significantly increased by both plumbing types to a similar level (52 to 104). Laboratory plumbing significantly changed the bacterial community, but toilet plumbing did not. Remarkably, water supply interruption/restoration led to serious water quality deterioration in both plumbing types but resulted in different changes. Physiochemically, discoloration was observed only in laboratory plumbing, along with sharp increases in Mn and Zn. Microbiologically, the increase in ATP was sharper in toilet plumbing than in laboratory plumbing. Some opportunistic pathogen-containing genera, e.g., Legionella spp. and Pseudomonas spp., were present in both plumbing types but only in disturbed samples. This study highlighted the esthetic, chemical, and microbiological risks associated with premise plumbing, for which system configuration plays an important role. Attention should be given to optimizing premise plumbing design for managing building water quality.

@en

Biofilm detachment contributes to water quality deterioration. However, the contributions of biofilm detachment from different pipes have not been quantified or compared. Following the introduction of partial reverse osmosis (RO) in drinking water production, this study analyzed particles at customers’ ends and tracked their origins to water distribution mains and service lines. For doing so, filter bags were installed in front of water meters to capture upstream detached particles, while biofilm from water main and service line were sampled by cutting pipe specimens. The results showed that elemental concentrations of the biofilm in mains were higher than those of service lines (54.3–268.5 vs. 27.1–44.4 μg/cm2), both dominated by Ca. Differently, filter bags were dominated by Fe/Mn (77.5–98.1%). After introducing RO, Ca significantly decreased in biofilms of mains but not service lines, but the released Fe/Mn rather than Ca arrived at customers’ ends. The ATP concentrations of service lines were higher than mains, which decreased on mains but increased in service lines after introducing RO. For the core ASVs, 13/24 were shared by service lines (17), mains (21), and filter bags (17), which were assigned mainly to Nitrospira spp., Methylomagnum spp., Methylocytis spp., and IheB2–23 spp. According to source tracking results, service lines contributed more than mains to the particulate material collected by filter bags (57.6 ± 13.2% vs. 13.0 ± 11.6%). To the best of our knowledge, the present study provides the first evidence of service lines’ direct and quantitative contributions to potential water quality deterioration at customers’ ends. This highlights the need for the appropriate management of long-neglected service line pipes, e.g., regarding material selection, length optimization, and proper regulation.@en

Vehicle proactive guidance strategies are used by ride-hailing platforms to mitigate supply–demand imbalance across regions by directing idle vehicles to high-demand regions before the demands are realized. This article presents a data-driven stochastic optimization framework for computing idle vehicle guidance strategies. The objective is to minimize drivers’ idle travel distance, riders’ wait time, and the oversupply costs (OSCs) and undersupply costs (USCs) of the platform. Specifically, we design a novel neural network that integrates gated recurrent units (GRUs) with mixture density networks (MDNs) to capture the spatial-temporal features of the rider demand distribution.@en

The transport and fate of nanoplastics (NPs) in aquatic environments are closely associated with their colloidal stability, which is affected by aging and natural organic matter (NOM) adsorption. This study systematically investigated the combined effects of photoaging and NOM (e.g. humic acids, HA; and a model protein, bovine serum albumin, BSA) on the aggregation kinetics of NPs (polystyrene, PS) in NaCl and CaCl2 solutions. Our results showed that photoaged NPs adsorbed less HA than pristine NPs due to weaker hydrophobic and π-π interactions. In return, HA showed weaker impacts on NPs’ stability after photoaging. Differently, photoaged NPs absorbed more BSA than pristine NPs due to stronger hydrogen bonding and electrostatic attraction. Thus, the inhibitory effects of BSA on the aggregation kinetics of NPs were enhanced after photoaging. Regarding the effects of NOM on the aging of NPs, our results showed that HA competed with NPs for photons and underwent photo-degradation. Subsequently, the destruction/reconstruction of adsorbed HA increased (in NaCl) or decreased (in CaCl2) the stability of NPs. Notably, light radiation-induced flocculation of BSA molecules, which wrapped and integrated NPs and lead to their destabilization. Overall, this study provided new insights into the aggregation behavior of NPs in aquatic systems, which have significant implications for predicting the transport and fate of NPs in complex real-world environments.@en

To reduce the vehicle relocation rate considering relieving disequilibrium of the supply-demand ratios across regions for car-sharing systems, in this paper, we propose a data-driven optimization framework by integrating the non-parametric learning algorithm and two-stage stochastic programming modeling technique to address the one-way station-based car-sharing relocation problem. In contrast with the most existing work that deals with demand uncertainty using predefined probability distributions, the learning-based framework is capable of handling demand uncertainty by learning the intrinsic pattern from large-scale historical data and computing high quality solutions. To validate the performance of our proposed approach, we conduct a group of numerical experiments based on New York taxicab trip record data set. The experimental results show that our proposed data-driven approach outperforms the parametric approaches and deterministic model in terms of business profit, relocation rate, and value of stochastic solution (VSS). Most significantly, compared with the deterministic approach, the vehicle relocation rates are reduced by approximate 80%, 70% and 40% under small fleet size, medium fleet size and large fleet size, respectively. In addition, the VSS of our approach is more than 3 times higher than the one of Poisson distribution by average.

@en

In ride-sharing services, travel time uncertainty significantly impacts the quality of matching solutions for both the drivers and the riders. This paper studies a one-to-many ride-sharing matching problem where travel time between locations is uncertain. The goal is to generate robust ride-sharing matching solutions that minimize the total driver detour cost and the number of unmatched riders. To this end, we formulate the ride-sharing matching problem as a robust vehicle routing problem with time window (RVRPTW). To effectively capture the travel time uncertainty, we propose a deep learning-based data-driven approach that can dynamically estimate the uncertainty sets of travel times. Given the NP-hard nature of the optimization problem, we design a hybrid meta-heuristic algorithm that can handle large-scale instances in a time-efficient manner. To evaluate the performance of the proposed method, we conduct a set of numeric experiments based on real traffic data. The results confirm that the proposed approach outperforms the non-data-driven one in several important performance metrics, including a proper balance between robustness and inclusiveness of the matching solution. Specifically, by applying the proposed data-driven approach, the matching solution violation rate can be reduced up to 85.8%, and the valid serving rate can be increased up to 42.3% compared to the non-data-driven benchmark.@en

Freelance drivers in ride-hailing systems may strategically accept or reject ride requests based on their projection of the profitability of the assigned rides. This driver acceptance uncertainty is mainly caused by the flat rate payment and the blind ride acceptance rule adopted by most ride-hailing platforms. As a result, a high driver rejection rate has been observed, causing a negative impact on the service quality and matching efficiency for the ride-hailing systems. In this paper, we propose a pricing mechanism to improve drivers’ average ride acceptance rate by offering personalized payments computed based on the characteristics of individual riders and the estimated acceptance rates of the drivers. Specifically, we model and predict the drivers’ ride acceptance rates through a binary choice model and incorporate it into the stochastic optimization problem for the ride-hailing system. This provides personalized payment for each driver in connection with the characteristics of the assigned ride and the preferences of the drivers. We then evaluate the performance of the proposed pricing mechanism through extensive numerical experiments based on RideAustin trip data from June 2016 to April 2017. The results suggest that our proposed pricing mechanism improves the drivers’ average acceptance rate by an average of 60% compared to some commonly used pricing schemes. It also significantly increases the platform’s expected profit and matching rate. This implies a strong potential for the proposed pricing mechanism to improve service reliability and quality in ride-hailing systems.@en

This paper proposes an integrated dispatching framework for matching drivers with riders in ride-hailing systems. The goal is to compute matching solutions that maximize social welfare and benefit both sides of the market, such that the sustainable growth of the ride-hailing system is ensured. The proposed framework integrates data-driven proactive guidance strategies with batched matching optimization to increase social welfare, improve matching rate and reduce rider wait time. Proactive guidance strategies are computed by leveraging short-term demand forecasts based on historical data. Taken the resulting guidance strategies as inputs, the batched matching algorithm computes optimal bipartite matching between drivers and riders in a batch. Using New York City taxi data from 2016 March 1st to March 31st as input, we conduct a numerical study to evaluate the performance of the proposed framework and compare it with existing approaches in the literature. Our results show that the proposed framework improves social welfare for up to 50%. It also increases the matching rate by an average of 20% and reduces the average rider wait time by over 15%. This implies a strong potential for the proposed dispatching framework to improve service quality in ride-hailing systems.@en

In this paper, we study a one-to-one matching ride-sharing problem to save the travellers' total travel time considering travel time uncertainty. Unlike the existing work where the uncertainty set is assumed to be known or roughly estimated, in this work, we propose a learning-based robust optimization framework to handle the issue properly. Specifically, we assume the travel time varies in an uncertainty set which is predicted by a machine learning approach- ARIMA using travel time historical data, the predicted uncertainty set then serves as the input parameter for the robust optimization model. To evaluate the proposed approach, we conduct a group of numerical experiments based on New York taxi trip record data sets. The results show that our proposed data-driven robust optimization approach outperforms the robust optimization model with a given uncertainty set in terms of total travel time savings. Further, the proposed approach can improve the travel time savings up to 112.8%, and 34% by average. Most importantly, our proposed approach is capable of handling the uncertainty in a more effective way when the uncertainty degrees become high.

@en

We propose a data-driven optimization model to reduce riders' wait time for vehicle guidance and rebalancing operations, considering the rider demands are under uncertainty. Instead of assuming a pre-defined rider demand distribution, we propose a data-driven framework that integrates Mixture Density Networks (MDNs) and a two-stage stochastic programming model. The integrated framework can compute high-quality guidance and rebalancing solutions that benefit drivers and riders in the ride-hailing system by leveraging the time-series historical data from real data sets. To prove the performance and effectiveness of our approach, we conduct a group of simulations based on the New York High Volume For-Hire Vehicle (HVFHV) trip records. The validation results show that the proposed method outperforms the data-driven deterministic models using GRU and moving average methods. Most significantly, the riders' average wait time using our proposed approach can be reduced by 75.9% compared to the batched matching mechanism.

@en

In this paper, we propose a data-driven robust optimization model to reduce total travel cost in ride-sharing systems under travel time uncertainty. Instead of using a pre-defined uncertainty set, we study a data-driven robust optimization approach that integrates gated recurrent units (GRUs) predictions with a one-stage robust optimization model. The proposed approach has the ability to compute high quality solutions by leveraging the large-scale historical data to derive the uncertainty set for the designed robust optimization model. To evaluate the proposed approach, we conduct a group of simulations based on the New York taxi trip record data sets. The validation results show that our data-driven robust optimization approach outperforms the robust optimization approach with a pre-defined uncertainty set in terms of travellers' total travel cost. Most importantly, the total travel cost under the proposed approach is reduced by up to 31.7%, and by 26.7% on average compared with the robust model with a pre-defined uncertainty set.

@en

We propose a learning-based approach for open driver guidance and rebalancing in ride-hailing platforms. The objective is to further enhance the wait time reduction benefit of batched matching by incorporating learning-based open driver guidance and rebalancing. By leveraging the rider demand data, the guidance solutions are computed through the integration of machine learning techniques with a two-stage stochastic programming model. To validate the performance of the proposed approach, we conduct numerical experiments using the New York taxi trip data sets. Our results show that the proposed approach outperforms the single value estimation model and the parametric model using Poisson distribution in terms of average wait time. When assuming the open drivers are randomly located before the batching time window, the proposed approach reduces more than 70% of average wait time compared to batched matching without guidance.

@en