Next Article in Journal
Adaptation Potential of Three Psychrotolerant Aquatic Bacteria in the Pan-Okhotsk Region
Next Article in Special Issue
Infrastructure Asset Management: Historic and Future Perspective for Tools, Risk Assessment, and Digitalization for Competence Building
Previous Article in Journal
Thermal Regime and Water Balance of Two Tropical High-Mountain Lakes in the Nevado de Toluca Volcano, Mexico
Previous Article in Special Issue
An Enabling Environment for Asset Management through Public Policy: The Benefits of Standardization and Application to the Water Sector
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Treatment Reliability-Based Method for Supporting Infrastructure Asset Management of Wastewater Treatment Plants

Urban Water Unit, Hydraulics and Environment Department, National Civil Engineering Laboratory, Av. Brasil 101, 1700-066 Lisbon, Portugal
*
Author to whom correspondence should be addressed.
Water 2022, 14(7), 1106; https://doi.org/10.3390/w14071106
Submission received: 15 February 2022 / Revised: 22 March 2022 / Accepted: 28 March 2022 / Published: 30 March 2022
(This article belongs to the Special Issue Infrastructure Asset Management of Urban Water Systems)

Abstract

:
A simple and consolidated reliability-based method widely used to unveil the real reliability and stability of wastewater treatment plants (WWTPs) is herein proposed to trigger decision making on operational improvements and asset management for maintaining or improving treatment effectiveness, reliability, and efficiency. Five-year data (2015–2019) from 16 Portuguese activated sludge WWTPs were used. For the 73% of the yearly data which fitted a lognormal distribution, Niku’s coefficient was computed to assess the plant annual reliability for biological oxygen demand (BOD5), chemical oxygen demand (COD), and total suspended solids (TSS). The standard deviation of the annual concentrations was used to characterize the plant stability, and the maximum standard deviations allowed to comply with the European discharge requirements for urban WWTPs were derived. The results demonstrate extended aeration WWTPs were more reliable and stable than conventional aeration WWTPs (0.98 reliability vs. 0.82 for BOD5, 0.97 vs. 0.91 for COD, and 0.94 vs. 0.89 for TSS). Furthermore, the lower reliabilities and stabilities were found for the smaller WWTPs. These results are important for strategic asset management for designing and rehabilitation of the wastewater treatment system. At tactical and operational levels, for resources’ allocation and operating conditions set up, the computed WWTP’s coefficient of variation allows establishing the mean effluent concentrations required for compliance with a given reliability for different scenarios of discharge requirements.

1. Introduction

Wastewater treatment plants (WWTPs) are being increasingly challenged to improve resources’ use efficiency and to provide higher levels of treatment to meet more stringent discharge consents and/or water reuse opportunities.
To manage or upgrade existing WWTPs or plan new ones, an important design factor is the reliability in meeting permit requirements. Reliability of a treatment plant may be defined as the probability of adequate performance for a specified period under specified conditions, i.e., the percent of the time that effluent concentrations meet specified permit requirements [1].
The need to continuously provide an effective and efficient service while infrastructures are ageing calls for increasingly sustainable infrastructure asset management (IAM) on strategic, tactical, and operational levels of planning [2]. Different approaches are used worldwide for business managers and accounts, water engineers, asset maintenance managers, and many elected officials, but the key role of performance metrics for IAM is consensual [3], as established in ISO 55000/55001/55002 standards on asset management [4,5,6]. Namely, the metrics are essential for diagnosing the performance in the status-quo scenario and for predicting it, considering different future scenarios, based on which measures/alternatives can be prioritized and results can be monitored [7,8].
Assessing the compliance with the discharge requirements is a rather complex process for WWTPs in EU Member States, since it requires the integration of a large volume of data and several criteria according to EU Directives 91/271/EEC [9] and 2000/60/EC [10]. To assist this process, we have developed a tool for a comprehensive assessment of treated wastewater quality that integrates performance indicators (PIs) and performance indices [11]. The indices tackle the plant reliability, i.e., they allow to easily compare the performance of different parameters over time and identify when the performance satisfied or failed the pre-established objectives and the distance remaining to achieve the targets set [11]. However, this assessment of reliability is not quantitative, and this feature limits its use as an IAM metric.
Herein, the novelty of the current work is to use a simple and consolidated reliability-based method to trigger decision making on operational improvements and asset management for maintaining or improving treatment effectiveness and efficiency. Such method allows to (i) diagnose the WWTP reliability, (ii) estimate it for different scenarios of discharge requirements, (iii) estimate the design/operating mean value of each parameter to meet the requirement with a given reliability, and (iv) derive the stability cut-off points to achieve the reliability needed for the compliance.
The use of probabilistic methods in setting discharge standards is a realistic and practical approach from an operational point of view [12]. When the plant effluent concentrations fit a lognormal distribution, the coefficient of reliability (COR) proposed by USEPA [13] is a simple and widely used method for reliability analysis [12,14]. To compare different WWTPs, a stability measure is needed, and the standard deviation is being used as the most appropriate stability indicator, as proposed by Niku et al. [13]. Other concentration distributions, e.g., Weibull or Gamma, require other probability models [15,16,17,18,19] or fault tree analysis. For example, fault tree analysis and Monte Carlo simulation has been used for mechanical reliability, to analyze risk of drinking water and, ultimately, for assessment of the violation of effluent biological oxygen demand (BOD5) from the standard limit for landscape irrigation to identify the causal failure [20].
The reliability approach based on Niku’s COR parameter [13] and on Silva et al.’s discharge compliance PI [11] is herein tested for 16 Portuguese WWTPs with activated sludge (AS) systems, the most widely used treatment around the world [21], with extended or conventional aeration regimes and different capacities.
Despite the high potential of this simple and consolidated method, to our knowledge, it has not been fully explored for supporting IAM.

2. Methods

2.1. WWTPs Analysed

Five-year (2015–2019) data of 16 Portuguese activated sludge WWTPs were used. As presented in Table 1, the WWTPs analyzed cover different capacities (763–54,000 m3/d) and treated volumes (446–38,974 m3/d 5-year medians), and two treatment sequences: (i) activated sludge after primary sedimentation, designed for conventional aeration (CAS) (not necessarily operated as so, if the plant is underutilized) and (ii) activated sludge without primary sedimentation, designed for extended aeration (EA). Namely, 7 CAS-WWTPs (with 5-year median inflows in the range of 4460–33,140 m3/d) and 9 EA-WWTPs (446–38,974 m3/d) were analyzed towards effluent BOD5, chemical oxygen demand (COD), and total suspended solids (TSS) compliance with the European discharge consents for urban wastewater treatment, namely, 25 mg/L BOD5, 125 mg/L COD, and 35 mg/L TSS (EU Directive 91/271/EEC).
The influent wastewater median concentrations varied from 165 mg/L to 550 mg/L BOD5 and from 311 mg/L to 983 mg/L COD depending on the industrial contribution. The percentile 25–75 range (P25–P75) of the BOD5 mass load to the reactor was 0.19–0.48 kg BOD5/m3·d. Figure 1 presents the boxplot results of capacity utilization (treated volume/design capacity ratio), hydraulic retention time (HRT), mixed-liquor suspended solids (MLSS), Food/Microorganisms ratio (F/M), and solids retention time (SRT) for the two AS-WWTP types. As expected, the analyzed EA-WWTPs presented higher HRT than the CAS-WWTPs (P25-P75 of 28-32 h vs. 7–21 h), higher MLSS (median 3710 mg/L vs. 3412 mg/L), higher SRT (16–23 d P25-P75 and 17 d median vs. 5–21 d P25-P75 and 9 d median), and lower F/M (0.06–0.09 d−1 P25-P75 and 0.07 d−1 median for EA-WWTPs vs. 0.12–0.39 d−1 P25–P75 and 0.18 d−1 median for CAS-WWTPs). Some CAS-WWTPs are underutilized (Figure 1), and therefore the operating conditions are closer to those typical of EA.
The Kolomogorov–Smirnov (K–S) test was used to verify if the concentration results of the 16 WWTPs analyzed in each year of the 5-year period fit a lognormal distribution at significant levels of 1 percent. Overall, more than 73% of the data fit a lognormal distribution (test results in Supplementary Materials, Tables S1–S3). Therefore, the Niku’s COR method based on the lognormality of the data could be used to estimate the reliability of these plants. Deviations from the lognormal distribution were mostly found for plant data series with many results below the limit of quantification (LOQ), i.e., with a plateau at LOQ.

2.2. Reliability Determination

The COR parameter developed by Niku et al. [13] was used to estimate the WWTP reliability for each parameter with minimum requirements set for discharge or reuse. In this method, the mean value (mx) is related to the standard (Xs) that must be achieved on a probability basis (Equation (1)):
mx = COR Xs
The coefficient of reliability is determined by Equation (2):
COR = [(Vx2 + 1)1/2] exp {− Z1−α [ln (Vx2 + 1)] ½}
where Vx is the coefficient of variation (standard deviation divided by mean), α is the probability of failure of meeting the standards, 1−α is the reliability level, and Z1α is the number of standard deviations away from the mean of a normal distribution.
Z1−α was computed by Equation (3):
Z 1 α   =   ln   [ m x / X s ( V x 2   +   1 ) ½ ] [ ln   ( V x 2   +   1 ) ] ½
and reliability 1−α is determined using the function NORM.S.DIST(Z1−α,TRUE) in excel.

2.3. Compliance Indicator

In line with the EU legislation for urban wastewater discharge, the PI “wtWQ03.2a, Compliance of discharged wastewater quality with Directive 91/271/EEC [%]” presented by Silva et al. in [11] involves the assessment of treated wastewater compliance with each parameter (Ji = 1, compliance; Ji = 0, no compliance), Equation (4):
wtWQ 03.2 a   =   i = 1 m J i m   ×   100
The determination of Ji integrates the several criteria defined in the directive, namely, for each parameter (BOD5, COD, TSS), the parametric values (Xs), the deviations allowed from the Xs, the minimum annual number of samples, and the maximum number of samples which are allowed to fail the Xs. A flowchart for a straightforward assessment of compliance is presented in Silva et al. [11].

3. Results and Discussion

3.1. Reliability vs. Compliance

Reliability and compliance were computed for BOD5, COD, and TSS, annually (2015 to 2019), for the 16 WWTPs analyzed whose yearly data fit the lognormal distribution. The results obtained are presented in Table 2, which also includes the number of samples and the mean values of the effluent concentration that were used to compute the 1−α values (the reliability), as explained in Section 2.2.
The reliability results were plotted against compliance, as shown in Figure 2, to establish the minimum reliability needed to comply with the EU directive discharge requirements, i.e., for ensuring Xs (Equation (1)) of 25 mg/L BOD5, 125 mg/L COD, and 35 mg/L TSS.
The aggregated results of all WWTPs analyzed during the 5-year period show the minimum reliability was 0.90 for BOD5 and COD and was 0.84 for TSS. These values are coherent with the maximum number of tests that are allowed to fail the Xs in relation with the minimum number of tests carried out. For example, when 12 tests are carried out, 2 tests are allowed to fail the Xs, which corresponds to 0.83 compliance; for 52 tests carried out, 0.90 compliance is required. However, a reliability value equal or above these cut-offs could result in a noncompliance if the maximum deviation is exceeded, as shown in Figure 2. The recommendation of Andraka and Dzienis for wastewater treatment plants under 50,000 equivalent population (1 EP = 60 g BOD5/d, EU Directive 91/271/EEC) is a minimum reliability of at least 0.94 for all parameters [16].

3.2. Reliability and Compliance vs. Stability

Stability is a measure of variation from the mean, and the standard deviation was herein used as the stability indicator [13]. Figure 3 displays the results of the 16 WWTPs analyzed during 2015–2019 with lognormal data distribution. The results show that higher standard deviations are associated with lower reliability, as expected, and aid in identifying the stability cut-off points to achieve the compliance.
For the pools of WWTPs studied, the maximum standard deviations of each parameter to achieve compliance are 6.4 mg/L for BOD5, 32 mg/L for COD, and 10.3 mg/L for TSS (Figure 3). These cut-off values must be read in Figure 3 (right) for no “compliance data” (above them, compliance may or may not be achieved).
These results are also consistent with Niku et al.’s conclusions [13], which found that plants with standard deviations greater than 10 mg/L for both BOD5 and TSS may be considered unstable.

3.3. Reliability and Stability of the Two Treatment Sequences Analyzed

Niku et al. [13] found reliability and stability of activated sludge processes to depend on the type of treatment, with step-aeration modification of activated sludge being more reliable and stable for BOD5 and conventional activated sludge being more reliable and stable for TSS.
Within our pool of 5-year results from 16 WWTPs, reliability and stability did not correlate with the AS operating conditions shown in Table 1. However, Welch F tests (used in the case of unequal variances) were conducted and showed a significant difference (p-values < 0.05) between the two AS clusters, CAS-WWTPs and EA-WWTPs, for both reliability and stability of BOD5 and COD effluent concentrations (results in Supplementary Materials, Table S4).
For the three parameters analyzed (BOD5, COD, and TSS), Figure 4 shows the EA-WWTPs (each number in the x-axis corresponds to a given WWTP each year) presented higher reliability (0.98, 0.97, and 0.94 mean values for BOD5, COD, and TSS in EA vs. 0.82, 0.91, and 0.89 for CAS) and higher stability (mean standard deviations of 4.7, 23.5, and 10.6 mg/L for BOD5, COD, and TSS in EA vs. 14.0, 37.8, and 14.3 mg/L for CAS). Niku et al. also present higher stability for extended aeration (5.3 mg/L for BOD5 and TSS) than for conventional activated sludge systems (9.5 mg/L for BOD5 and 16 mg/L for TSS) [13].

3.4. Reliability and Stability vs. Treatment Capacities

Wastewater flowrate affects most operating conditions determining the treatment effectiveness (e.g., detention times, loads) and efficiency (e.g., unit energy consumption [22]). Therefore, the effect of the treated wastewater volume on treatment reliability and stability was analyzed.
Figure 5 shows no linear correlation between treatment reliability and treated volume, but WWTPs treating more than 15,000 m3/d (EA-WWTPs) and particularly more than 20,000 m3/d (EA-WWTPs and CAS-WWTPs) were more reliable (>0.90) and stable. These results agree with literature. Niku et al. [13] found no relationship between plant size and stability. However, Bunce et al. [14] reported that the smallest WWTPs appeared to be less stable than the slightly larger WWTPs across all technology types.
These results are important for strategic asset management concerning the designing and rehabilitation of the wastewater treatment system in terms of number of plants, their capacity, and treatment sequence. For example, after this study, the water utility of the underutilized CAS-WWTP C (Table 1 and Table 2) decided to decommission the primary sedimentation and to properly operate the plant as an EA-WWTP.

3.5. Estimating the Target Effluent Mean Values for a Given Reliability

After characterizing the plant reliability, namely its coefficient of variation (Vx in Equation (2)), one can estimate the target mean value (mx, in Equation (1)) that should have been met in each year for achieving the desired objective (Xs) of each parameter with a given reliability. The higher the latter, the lower (more stringent) the target mean value is. The target mean values for BOD5, COD, and TSS were estimated for reliability values of 0.85, 0.9, 0.95, and 0.99. The results aggregated per AS type are show in Figure 6 and Figure 7, and those of each WWTP are presented in Table 3 for 0.85 and 0.95 reliability.
Assuming similar performance and influent scenarios are likely to occur, this back casting supports establishing the target mean values for the daily WWTP operation. For example, for WWTP P with a coefficient of variation (Vx) of 0.3–0.5 of the BOD5 (5-year range), changing the reliability from 0.85 to 0.95 implies a reduction of target mean values from 17.4–19.0 mg/L to 13.3–15.7 mg/L BOD5. Furthermore, this back casting step allows the assessment of the impact of a reliability change in the discharge requirements and the feasibility of its compliance, which triggers the rehabilitation measures.
The aggregated results of the EA-WWTPs (Figure 6) and of the CAS-WWTPs (Figure 7) can be used as reference ranges for WWTPs within each cluster diversity (influent characteristics, capacities, and operating conditions) for plant design or performance benchmarking. For example, for 0.95 reliability, the medians of target mean values for the nine EA-WWTPs studied were 13.8 mg/L BOD5, 77.3 mg/L COD, and 16.7 mg/L TSS; for the seven CAS-WWTPs studied, they were 12.8 mg/L BOD5, 69.6 mg/L COD, and 17.2 mg/L TSS. The higher the AS-specific coefficient of variation (higher for CAS than for EA), the lower the design values must be. Niku et al. proposed, as recommended design values of CAS-WWTPs, 14.5 mg/L for effluent BOD5 and 11.6 mg/L for effluent TSS [13].
The target mean values (mx) for 0.85 reliability (close to the reliability cut-off for compliance with the EU minimum requirements for urban WWTP discharge when 12 tests are carried out) are significantly lower than the EU minimum requirements (P25 of mx of all WWTPs varied from 60% to 70% of Xs, depending on the parameter). This behavior was observed to a larger extent for the CAS-WWTPs than for the EA-WWTPs (P25 values for 0.85 reliability in Figure 6 and Figure 7, namely 15.1 mg/L BOD5, 81.2 mg/L COD, and 21.2 mg/L TSS for CAS-WWTPs vs. 17.3 mg/L BOD5, 87.5 mg/L COD, and 22.7 mg/L TSS for EA-WWTPs).
These are important data to consider in the IAM plans (strategic, tactic, and operation) for treatment type and capacity selection, resources allocation, and operating conditions set up.
Taheriyoun and Moradinejad [20] identified the human factor as a priority to improve the reliability of the plant, along with complementary actions such as the automation level increase. The same lesson was learned within the iEQTA project, although no quantitative assessment of reliability vs. human resources (number and skills) was conducted.

4. Conclusions

The simple and consolidated Niku’s reliability-based method was herein integrated with the assessment of the compliance of urban WWTPs with discharge requirements. This integrated analysis allows to estimate the WWTP reliability, stability, compliance and target effluent mean values for a given reliability, which is key information to trigger decision making on operational improvements and asset management (new investment, rehabilitation, or retrofitting).
The results obtained demonstrate that the nine EA-WWTPs were significantly more reliable and stable than the seven CAS-WWTPs analyzed. In addition, EA-WWTPs treating more than 15,000 m3/d and EA and CAS-WWTPs treating more than 20,000 m3/d are more reliable (>0.90) and stable.
The results support the tactical and operational levels of IAM (resources’ allocation and operating conditions) by estimating, for different scenarios of discharge requirements, the WWTP reliability target, the corresponding effluent mean values, and the stability cut-off point (standard deviations). On a strategic level of IAM, the results can be used as reference ranges for WWTPs within each cluster diversity (influent characteristics, capacities, and operating conditions) for plant design or performance benchmarking.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/w14071106/s1, Table S1: Kolomogorov-Smirnov (K-S) test of BOD5 for each WWTP in each year; Table S2: Kolomogorov-Smirnov (K-S) test of COD for each WWTP in each year; Table S3: Kolomogorov-Smirnov (K-S) test of TSS for each WWTP in each year; Table S4: Welch F tests of the two AS clusters, CAS-WWTPs and EA-WWTPs, for reliability and stability.

Author Contributions

Conceptualization, C.S. and M.J.R.; methodology, C.S. and M.J.R.; software, C.S.; validation, C.S. and M.J.R.; formal analysis, C.S. and M.J.R.; investigation, C.S. and M.J.R.; resources, C.S. and M.J.R.; data curation, C.S.; writing—original draft preparation, C.S. and M.J.R.; writing—review and editing, C.S. and M.J.R.; visualization, C.S.; supervision, M.J.R.; project administration, C.S.; funding acquisition, C.S. and M.J.R. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Acknowledgments

The authors acknowledge the Portuguese water utilities for providing the data.

Conflicts of Interest

The authors declare no conflict of interest.

Abbreviations

ASActivated Sludge
BOD5Biological Oxygen Demand
CORCoefficient of Reliability
CASConventional Aeration
CODChemical Oxygen Demand
EAExtended Aeration
F/MFood/Microorganisms ratio
HRTHydraulic Retention Time
IAMInfrastructure Asset Management
LOQlimit of quantification
MLSSMixed-Liquor Suspended Solids
PIsPerformance Indicators
SRTsolids retention time
TSSTotal Suspended Solids
WWTPsWastewater Treatment Plants

References

  1. Metcalf and Eddy Ltd. Wastewater Engineering—Treatment and Resource Recovery, 5th ed.; Tchobanoglous, G., Stensel, H.D., Tsuchihashi, R., Burton, F., Eds.; McGraw-Hill Education: New York, NY, USA, 2014. [Google Scholar]
  2. Cardoso, M.A.; Poças, A.; Silva, M.S.; Ribeiro, R.; Almeida, M.C.; Brito, R.S.; Coelho, S.T.; Alegre, H. Innovation results of IAM planning in urban water services. Water Sci. Technol. 2016, 74, 1518–1526. [Google Scholar] [CrossRef] [PubMed]
  3. Alegre, H.; Covas, D.; Coelho, S.T.; Almeida, M.C.; Cardoso, M.A. Integrated Approach for Infrastructure Asset Management of Urban Water Systems. In Proceedings of the IWA 4th LESAM, Mülheim an der Ruhr, Germany, 27–30 September 2011. [Google Scholar]
  4. ISO 55000:2014; Asset Management—Overview, Principles, and Terminology. ISO: Geneva, Switzerland, 2014.
  5. ISO 55001:2014; Asset Management—Management Systems—Requirements. ISO: Geneva, Switzerland, 2014.
  6. ISO 55002:2014; Asset Management—Management Systems—Guidelines for the Application of ISO 55001. ISO: Geneva, Switzerland, 2014.
  7. Loureiro, D.; Silva, C.; Cardoso, M.A.; Mamade, A.; Alegre, H.; Rosa, M.J. The Development of a Framework for Assessing the Energy Efficiency in Urban Water Systems and Its Demonstration in the Portuguese Water Sector. Water 2020, 12, 134. [Google Scholar] [CrossRef] [Green Version]
  8. Silva, C.; Rosa, M.J. Performance assessment of 23 wastewater treatment plants—A case study. Urban Water J. 2020, 17, 78–85. [Google Scholar] [CrossRef]
  9. EC. Council Directive of 21 May 1991 concerning urban waste-water treatment (91/271/EEC). Off. J. Eur. Union 1991, 135, 40–52. [Google Scholar]
  10. EC. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy. Off. J. Eur. Parliam. 2000, L327, 1–82. [Google Scholar]
  11. Silva, C.; Quadros, S.; Ramalho, P.; Rosa, M.J. A tool for assessing treated wastewater quality in urban WWTPs. J. Environ. Manag. 2014, 146, 400–406. [Google Scholar] [CrossRef] [PubMed]
  12. Oliveira, S.C.; Von Sperling, N. Reliability analysis of wastewater treatment plants. Water Res. 2008, 42, 1182–1194. [Google Scholar] [CrossRef] [PubMed]
  13. Niku, S.; Schroeder, E.D.; Tchobanoglous, G.; Samaniego, F.J. Performance of Activated Sludge Processes: Reliability, Stability and Variability; EPA/600/2-81/227; U.S. Environmental Protection Agency: Washington, DC, USA, 1981. [Google Scholar]
  14. Bunce, J.T.; Graham, D.W. A Simple Approach to Predicting the Reliability of Small Wastewater Treatment Plants. Water 2019, 11, 2397. [Google Scholar] [CrossRef] [Green Version]
  15. Andraka, D. Reliability Evaluation of Wastewater Treatment Plant Impact on the Receiving Waters. J. Ecol. Eng. 2019, 20, 226–231. [Google Scholar] [CrossRef]
  16. Kurek, K.; Bugajski, P.; Operacz, A.; Śliz, P.; Jóźwiakowski, K.; Almeida, A. Reliability assessment of pollution removal of wastewater treatment plant using the method of Weibull. E3S Web Conf. 2020, 171, 01007. [Google Scholar] [CrossRef]
  17. Jóźwiakowska, K.; Marzec, M. Efficiency and reliability of sewage purification in long-term exploitation of the municipal wastewater treatment plant with activated sludge and hydroponic system. Arch. Environ. Prot. 2020, 46, 30–41. [Google Scholar]
  18. Makowska, M.; Spychała, M.; Pawlak, M. Efficacy and reliability of wastewater treatment technology in small meat plants. Desalin. Water Treat. 2021, 221, 1–10. [Google Scholar] [CrossRef]
  19. Zawadzka, B.; Siwiec, T.; Marzec, M. Effectiveness of Dairy and Domestic Wastewater Treatment and Technological Reliability of the Wastewater Treatment Plant in Michów, Poland. J. Ecol. Eng. 2021, 22, 141–151. [Google Scholar] [CrossRef]
  20. Taheriyoun, M.; Moradinejad, S. Reliability analysis of a wastewater treatment plant using fault tree analysis and Monte Carlo simulation. Environ. Monit. Assess. 2015, 187, 4186. [Google Scholar] [CrossRef] [PubMed]
  21. Wu, L.; Ning, D.; Zhang, B.; Li, Y.; Zhang, P.; Shan, X.; Zhang, Q.; Brown, M.; Li, Z.; Van Nostrand, J.; et al. Global diversity and biogeography of bacterial communities in wastewater treatment plants. Nat. Microbiol. 2019, 4, 1183. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  22. Silva, C.; Rosa, M.J. Energy performance indicators of wastewater treatment—A field study with 17 Portuguese plants. Water Sci. Technol. 2015, 72, 510–519. [Google Scholar] [CrossRef] [PubMed]
Figure 1. Boxplots of the operating conditions for the two clusters of AS-WWTPs (5-year data, 9 CAS-WWTPs and 7 EA-WWTPs).
Figure 1. Boxplots of the operating conditions for the two clusters of AS-WWTPs (5-year data, 9 CAS-WWTPs and 7 EA-WWTPs).
Water 14 01106 g001
Figure 2. Plant reliability vs. compliance with the EU discharge requirements for urban WWTPs towards BOD5, COD, and TSS (5-year data, 16 activated sludge WWTPs).
Figure 2. Plant reliability vs. compliance with the EU discharge requirements for urban WWTPs towards BOD5, COD, and TSS (5-year data, 16 activated sludge WWTPs).
Water 14 01106 g002
Figure 3. Plant reliability vs. standard deviation and standard deviation vs. compliance with EU discharge requirements for urban WWTPs towards BOD5, COD, and TSS (5-year data, 16 activated sludge WWTPs).
Figure 3. Plant reliability vs. standard deviation and standard deviation vs. compliance with EU discharge requirements for urban WWTPs towards BOD5, COD, and TSS (5-year data, 16 activated sludge WWTPs).
Water 14 01106 g003
Figure 4. Plant annual reliability and standard deviation of BOD5, COD, and TSS effluent concentrations (each number in the x-axis corresponds to a given WWTP each year; 5-year data, 7 CAS-WWTPs and 9 EA-WWTPs).
Figure 4. Plant annual reliability and standard deviation of BOD5, COD, and TSS effluent concentrations (each number in the x-axis corresponds to a given WWTP each year; 5-year data, 7 CAS-WWTPs and 9 EA-WWTPs).
Water 14 01106 g004
Figure 5. Plant annual reliability and standard deviation of BOD5, COD, and TSS effluent concentrations vs. treated wastewater (5-year data, 9 EA-WWTPs and 7 CAS-WWTPs).
Figure 5. Plant annual reliability and standard deviation of BOD5, COD, and TSS effluent concentrations vs. treated wastewater (5-year data, 9 EA-WWTPs and 7 CAS-WWTPs).
Water 14 01106 g005
Figure 6. Target mean value (mx) of BOD5, COD, and TSS effluent concentrations for achieving different reliability targets in 9 EA-WWTPs (5-year data).
Figure 6. Target mean value (mx) of BOD5, COD, and TSS effluent concentrations for achieving different reliability targets in 9 EA-WWTPs (5-year data).
Water 14 01106 g006
Figure 7. Target mean value (mx) of BOD5, COD, and TSS effluent concentrations for achieving different reliability targets in 7 CAS-WWTPs based on 5-year data.
Figure 7. Target mean value (mx) of BOD5, COD, and TSS effluent concentrations for achieving different reliability targets in 7 CAS-WWTPs based on 5-year data.
Water 14 01106 g007
Table 1. Treatment type, designs capacity, and operating conditions of the 16 WWTPs analyzed. Values shown are 5-year medians (P50) and P25–P75 are between brackets.
Table 1. Treatment type, designs capacity, and operating conditions of the 16 WWTPs analyzed. Values shown are 5-year medians (P50) and P25–P75 are between brackets.
WWTPAS Design
Capacity
(103 m3/d)
Treated Wastewater
(103 m3/d)
Influent BOD5 (mg/L)Influent COD (mg/L)HRT in
Reactor (h)
MLSS (mg/L)SRT (d)F/M (d−1)
ACAS 9.54.5
(3.8–5.5)
420
(320–540)
729
(509–980)
23
(18–27)
3412
(2590–4140)
38
(15–51)
0.09
(0.07–0.13)
BCAS 4.45.9
(4.8–7.5)
268
(187–377)
528
(325–711)
6.9
(5.5–8.5)
2840
(2250–3690)
5.7
(3.7–8.3)
0.25
(0.16–0.42)
CCAS 42.911.6
(9.9–13.6)
500
(390–640)
900
(714–1100)
21
(18–25)
3428
(2873–4086)
21
(16–32)
0.13
(0.09–0.16)
DCAS 27.912.8
(11.9–14.0)
490
(370–590)
983
(840–1113)
14
(13–26)
4770
(3870–5683)
21
(13–35)
0.13
(0.11–0.17)
ECAS 26.017.5
(16.3–18.9)
331
(250–412)
560
(415–707)
5.8
(5.4–6.3)
1100
(800–1500)
1.8
(1.1–2.6)
0.81
(0.56–2.6)
FCAS 18.422.0
(17.3–26.9)
225
(164–303)
501
(365–692)
7.2
(5.9–9.2)
3000
(2610–3507)
4.7
(3.9–5.9)
0.23
(0.17–0.32)
GCAS 54.033.1
(30.8–35.4)
340
(250–560)
642
(491–893)
11.4
(10.6–12.3)
3580
(2915–4815)
8.9
(6.8–11.5)
-
HEA1.20.45
(0.39–0.54)
258
(176–346)
520
(320–700)
39
(32–45)
2240
(1750–2713)
56
(50–74)
0.13
(0.08–0.18)
IEA0.760.78
(0.61–1.2)
208
(107–358)
459
(293–760)
32
(22–43)
3625
(2600–5100)
17
(12–27)
0.05
(0.03–0.09)
JEA11.48.6
(6.9–10.8)
165
(110–240)
311
(234–483)
32
(26–40)
2613
(2048–3398)
17
(11–25)
0.07
(0.05–0.09)
KEA15.110.6
(8.6–12.7)
320
(277–355)
924
(758–1059)
32
(26–39)
4218
(3910–4586)
17 (17–20)0.07
(0.06–0.09)
LEA28.116.0
(12.5–20.7)
283
(232–326)
813
(669–956)
28
(20–37)
3375
(3100–3640)
14
(9–17)
0.10
(0.07–0.13)
MEA35.919.9
(16.6–23.3)
550
(371–630)
938
(730–1064)
19
(16–22)
3710
(3433–4085)
14
(11–17)
-
NEA25.621.3
(16.5–25.4)
320
(277–355)
924
(758–1059)
31
(26–40)
4890
(4505–5200)
21
(19–24)
0.07
(0.06–0.08)
OEA24.922.2
(14.9–27.4)
246
(189–296)
772
(575–925)
27
(22–39)
4280
(3879–4650)
19
(16–23)
0.06
(0.05–0.08)
PEA44.339.0
(32.0–45.5)
335
(243–398)
871
(587–1134)
30
(25–37)
5270
(4595–5920)
24
(21–28)
0.06
(0.04–0.08)
CAS: activated sludge system downstream from primary sedimentation, EA: activated sludge without primary sedimentation.
Table 2. WWTP reliability and compliance (Ji) for BOD5, COD, and TSS (“-“data do not fit a lognormal distribution).
Table 2. WWTP reliability and compliance (Ji) for BOD5, COD, and TSS (“-“data do not fit a lognormal distribution).
WWTPYearTreated
Volume (m3/d)
BOD5CODTSS
No.
Samples
mx (mg/L) Ji1−αNo.
Samples
mx (mg/L) Ji1−αNo.
Samples
mx (mg/L) Ji1−α
A201546065218.7000.855269.9010.985218.0010.96
A201657956616.3010.9237460.4010.996519.7010.96
A201743286317.9010.9036373.3010.986319.9010.94
A201851786319.6300.7637585.5910.896422.4810.89
A201949436714.3310.9336964.4710.986616.3310.96
B201562515324.420-10586.2200.8210533.200-
B201658525118.9200.8010268.4310.9010225.550-
B201752175328.3200.47105100.8000.7610733.900-
B201865875322.7400.6710566.4510.9410525.271-
B2019650710718.791-10761.0010.9710724.471-
C201511,16410312.9110.9510361.6511.0010317.9610.98
C201612,71810411.991-10452.9110.9910414.931-
C201712,09810021.9900.7010184.3200.9010020.5610.92
D201512,3585123.3700.6836118.1200.6612530.4500.71
D201612,9355218.5600.7826101.4200.755216.0010.95
D201712,2645227.0400.732688.8500.835321.5800.86
D201813,7398839.9200.4926124.2700.6311345.4800.55
D201914,3872425.4200.6626137.0000.602421.3800.85
E201516,5978315.5000.868350.6010.998327.3000.76
E201617,0697810.1010.987830.0011.007810.4000.95
E201717,0778321.9000.728444.8010.988320.7000.85
E201817,3429216.8900.819155.7800.949215.2900.92
E201920,7759214.2600.869254.6810.999111.7110.97
F201521,57913914.1610.9434973.041-14118.7610.95
G201532,0607012.6910.997053.6011.007020.3300.92
G201634,0137311.3610.987349.0811.007320.5800.90
G201733,0736810.9610.986852.3811.006817.3210.93
G201833,1537612.7110.967761.8810.987719.4210.93
G201932,8106511.9510.977652.4011.007621.3010.94
H20174336616.000-6640.2011.007015.501-
H20185584010.6700.944335.2610.99438.5810.99
H2019406336.2711.003335.1811.00335.3611.00
I201615822512.831-2446.421-2416.9610.94
I20177332430.920-2499.0000.762435.7100.68
I20189282415.790-2461.1010.912421.8810.84
I20199471218.3300.821267.3010.971222.7310.92
J201582969715.660-9872.0310.939822.5900.85
J201611,07911816.2900.8311865.8410.9411824.4700.82
J2017872910412.6500.9110455.1811.0010414.9510.98
J2018934110515.700-10562.9010.9410518.3910.90
J20198968988.0911.009852.9910.999813.2810.98
K20159710606.621-24157.3111.002419.071-
K201610,7913510.7211.0024459.9511.0024415.3410.98
K20179866349.5311.0024170.6611.0024118.7210.95
K201811,2183510.8910.9923666.6610.9823619.3610.94
K201911,0533310.7911.0023268.7610.9722418.7410.93
L201515,244607.2811.0024180.681-2416.791-
L201816,824367.4211.0023781.821-2376.921-
L201917,826367.3611.0024379.8910.982436.421-
M201516,201665.891-6633.2711.00667.3411.00
M201623,799697.9511.006938.1211.006912.0110.99
M201718,212567.8611.007237.2911.007211.2611.00
M201823,462669.6810.986851.0110.996819.1700.90
M201924,9525410.5010.976050.4010.996017.7510.96
N201520,813606.581-24159.6811.002419.421-
N201624,5263511.5110.9824470.7810.9824415.6010.96
N201719,378349.1211.0024369.2811.0024314.2310.97
N201820,8883512.3110.9721674.8210.9721018.8010.92
N201920,9433413.5010.9521076.6610.9521020.7410.91
O201620,845348.0011.0024457.9211.002448.111-
O201717,631347.2911.0024166.841-2419.371-
O201823,053368.8610.9923666.581-2367.981-
O201924,009357.7411.0024158.871-2416.561-
P201535,7925911.4911.0024365.4710.9824312.6810.98
P201639,5033410.8810.9924355.6010.9824313.1510.98
P201735,656349.3811.0024159.2411.0024112.6310.99
P201839,075359.8010.9923751.101-2379.771-
P201940,380358.8611.0024152.7611.002418.951-
Table 3. BOD5, COD, and TSS target mean values for each WWTP in each year, for 0.85 and 0.95 reliability.
Table 3. BOD5, COD, and TSS target mean values for each WWTP in each year, for 0.85 and 0.95 reliability.
WWTPYearBOD5CODTSS
Vx mx (mg/L)Vx mx (mg/L)Vx mx (mg/L)
1−α = 0.85 1−α = 0.951−α = 0.85 1−α = 0.951−α = 0.85 1−α = 0.95
A20150.3418.715.30.2996.981.60.4524.618.9
A20160.3618.515.00.3195.679.50.3825.620.5
A20170.3119.115.90.2996.981.60.4524.618.9
A20180.5716.612.00.3692.274.40.4624.418.7
A20190.4817.313.10.3592.975.40.5223.717.6
B2015---0.6381.257.1---
B20160.4617.513.40.6680.455.8---
B20170.4517.613.50.4587.767.4---
B20180.6016.411.70.5185.263.5---
B2019---0.4588.067.8---
C20150.4817.313.10.2996.781.20.3825.620.5
C2016---0.4189.970.8---
C20170.6216.311.50.3891.473.20.4724.318.5
D20150.9415.09.30.7478.452.40.9121.213.2
D20161.0214.98.91.1373.742.40.623.016.4
D20172.2615.36.80.5085.564.00.6922.315.2
D20181.2114.78.20.7578.151.91.0720.712.2
D20191.0214.98.90.8776.148.20.8321.513.8
E20150.6616.111.10.4986.164.90.9820.912.7
E20160.5316.912.50.2798.483.71.6720.610.2
E20170.9815.09.10.6381.157.11.0220.812.5
E20180.9315.19.30.7079.253.91.0520.812.3
E20190.9415.09.20.4488.368.30.8821.313.4
F20150.4417.713.7---0.4624.518.7
G20150.3019.316.10.22102.289.60.523.917.9
G20160.4517.613.50.3195.579.30.5523.517.2
G20170.4617.413.30.2997.181.80.6522.615.8
G20180.4817.313.20.3792.074.00.5123.917.9
G20190.4717.413.20.3394.377.50.3925.520.3
H2017---0.5185.263.6---
H20180.8515.39.70.6680.455.90.8521.413.6
H20190.5217.012.60.21102.790.40.7422.014.7
I2016------0.5923.116.5
I2017---1.0174.544.81.2220.511.5
I2018---0.8177.050.00.9321.113.1
I20190.4917.213.00.3891.272.80.3526.021.1
J2015---0.4687.266.70.6322.716.0
J20160.7915.510.10.5285.063.20.6122.916.3
J20170.8115.410.00.3096.180.20.5323.717.5
J2018---0.5882.659.50.7921.714.2
J20190.4717.413.20.3891.373.00.5523.517.2
K2015---0.3195.579.4---
K20160.3618.515.00.2698.984.50.4624.518.8
K20170.3019.316.10.21103.090.70.4624.518.7
K20180.3818.214.60.3493.876.70.4724.418.6
K20190.3418.715.30.3692.574.70.5323.717.5
L20150.3318.915.6------
L20180.2619.917.0------
L20190.2619.816.90.23101.788.8---
M2015 0.19104.893.60.5723.216.8
M20160.4617.513.40.2798.684.10.5223.817.7
M20170.4217.813.90.23101.588.50.5123.917.8
M20180.5916.511.80.4488.268.10.722.215.1
M20190.5916.511.80.4786.866.00.4824.318.4
N2015---0.3195.579.4---
N20160.4417.713.70.3096.380.50.6222.816.1
N20170.4317.813.90.23101.588.40.5923.016.5
N20180.4317.713.80.2996.781.10.5823.116.7
N20190.4717.413.20.3394.577.80.5123.817.8
O20160.3418.715.30.23101.087.7---
O20170.3219.015.6------
O20180.4717.313.2------
O20190.3218.915.6------
P20150.3219.015.70.3493.776.60.6322.716.0
P20160.4317.813.80.4488.568.70.6422.615.9
P20170.3518.615.10.3195.879.90.5123.917.9
P20180.4617.413.3------
P20190.4018.114.30.3792.274.3---
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Silva, C.; Rosa, M.J. A Treatment Reliability-Based Method for Supporting Infrastructure Asset Management of Wastewater Treatment Plants. Water 2022, 14, 1106. https://doi.org/10.3390/w14071106

AMA Style

Silva C, Rosa MJ. A Treatment Reliability-Based Method for Supporting Infrastructure Asset Management of Wastewater Treatment Plants. Water. 2022; 14(7):1106. https://doi.org/10.3390/w14071106

Chicago/Turabian Style

Silva, Catarina, and Maria João Rosa. 2022. "A Treatment Reliability-Based Method for Supporting Infrastructure Asset Management of Wastewater Treatment Plants" Water 14, no. 7: 1106. https://doi.org/10.3390/w14071106

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop