House of Quality Planning Matrix for Evaluating Wastewater Nutrient Management Technologies at Three Scales Within a Sewershed

Abstract

Nutrients (nitrogen and phosphorus) can be removed or recovered from wastewater at different points in the collection and treatment process, for example, from individual buildings or households, from the conveyance (sewer) system, or at the centralized treatment plant. Multiple technologies are available for nutrient removal and recovery at any of these points of application, but the appropriateness of a candidate technology depends on the scale of application; a technology that is appropriate for nutrient removal/recovery at a large centralized treatment plant may not be appropriate at an individual household. Hence, a need exists for a tool that enables municipalities to select scale- and context-specific technologies for nutrient management. Accordingly, the objectives of this article are (1) to review nutrient management technologies that are available at the building, community, and city scales and (2) to develop a planning matrix that evaluates the appropriateness of nutrient management technologies at these three scales based on ten practical characteristics. The planning matrix developed herein is similar in structure to a house of quality (HoQ), which is a quality function deployment method typically used in commercial businesses to determine how well a product meets the needs of its customers. Because each sewershed (i.e., the area and population served by a treatment plant and its associated sewer network) is unique, the planning matrix can be customized to determine the most appropriate nutrient management technologies for any given municipality and therefore represents a flexible tool for sewershed-scale nutrient management. We apply the new planning matrix to identify emerging nutrient recovery technologies (e.g., ion exchange, chemical precipitation, and membrane bioreactors) that may be preferable alternatives to current baseline technologies. At each scale, the technologies ranked as best by the HoQ method are consistent with those commonly employed in practice at present. However, future trends will likely affect technologies, weightings, and scores and therefore change the ranking of the technologies.

Introduction

A typical municipal wastewater treatment system consists of a centralized treatment plant that receives raw wastewater from a network of sewer pipes, which in turn collect wastewater from individual homes and businesses throughout the municipality. The geographic area and populace served by such a network are known as the sewershed of the sanitation system (Heidler et al., 2006; Teerlink et al., 2012). Just as a watershed is a geographic area from which all surface runoff drains through a single point of exit, a sewershed is drained as the final effluent from the centralized treatment plant. The sewershed and its infrastructure can be divided into three distinct scales: the smallest scale is individual homes or businesses at which wastewater is generated (building scale); the intermediate scale is the conveyance (sewer) system of a particular neighborhood or subregion of the overall sewershed (community scale); and the largest scale (city scale) is the centralized treatment plant and the sewershed that ultimately drains through that centralized plant.

Historically, the main purposes of such sanitation systems have been twofold: to protect human health by minimizing human contact with fecal pathogens and to protect environmental and ecological health by removing chemicals of concern before discharging wastewater into a receiving surface water. Constituents of concern include dissolved organic carbon (which exerts a biochemical oxygen demand in receiving waters) and the nutrients nitrogen and phosphorus (which can cause algal blooms and eutrophication in receiving waters). In most cases, any treatment of wastewater has been performed solely at the centralized treatment plant, rather than at the point of generation or in the conveyance system, and has consisted solely of removal of regulated constituents rather than recovery of those constituents in a useful form.

Recently, however, an important paradigm shift has been in progress, and municipal wastewater is increasingly viewed not merely as a waste product that requires treatment but also as a resource from which valuable products can be derived (e.g., Guest et al., 2009; Mo and Zhang, 2013; Water Environment Federation [WEF], 2014; Englehardt et al., 2016). Associated with that paradigm shift is a renewed interest in the possibility of on-site or decentralized treatment systems, that is, treatment systems that would be deployed at either the building scale or the community scale rather than at the city scale (Crites and Tchobanoglous, 1998; Tchobanoglous et al., 2004; Gikas and Tchobanoglous, 2009; Verstraete et al., 2009; The Johnson Foundation at Wingspread, 2014; Roefs et al., 2017). Technologies across all three scales can be combined to form a distributed system. Therefore, the term sewershed not only encompasses buildings that are currently on sewer, but may move to a decentralized treatment system in the future, but also buildings that could reasonably be connected to a sewer, but currently use decentralized technologies that may not necessarily include a sewer (e.g., a suburban home using a septic system) (Water Environment Research Foundation [WERF], 2011). Therefore, two important and interrelated questions have recently arisen: First, can nitrogen and/or phosphorus be economically recovered from wastewater in a useful form rather than merely removed from wastewater and discarded? Second, how should we best manage nitrogen and phosphorus at all three scales, not just at the city scale?

Many technologies already exist to remove and/or recover nitrogen and phosphorus from wastewater and many others have been proposed or are in different stages of technical development and maturity (Yeoman et al., 1988; Morse et al., 1998; Doyle and Parsons, 2002; de-Bashan and Bashan, 2004; Cai et al., 2013; Water Environment Federation [WEF], 2014). However, it must be recognized that the applicability or appropriateness of a candidate nutrient removal/recovery technology depends on the scale at which it is to be deployed. A technology that is appropriate for nutrient removal/recovery at a large centralized treatment plant, for instance, may not be appropriate at an individual household. This scale dependence of candidate technologies arises from a number of considerations, such as availability of space, esthetics, training, or expertise of the system operators (e.g., a licensed wastewater treatment plant [WWTP] operator versus an individual homeowner), and economies of scale. Furthermore, different municipalities or utilities (or even different sewersheds within a single municipality) may have different priorities due to a number of possible factors—different regulatory constraints, different concerns of the citizens, or stakeholders in the municipality, etc. Therefore, what is needed is a systematic framework or methodology for assessing candidate nutrient management technologies that account for scale of application. That framework should also be flexible enough to account for differing needs or priorities of different municipalities or utilities. There have been a few previous efforts toward this objective through life cycle assessment of nutrient management technologies (e.g., Lundin et al., 2000; Cornejo et al., 2016); however, life cycle analysis prioritizes environmental sustainability without accounting for certain practical factors (cost, esthetics, technical maturity, etc.) that are essential to most stakeholders and decision makers.

Therefore, the objectives of this article are (1) to review nutrient management technologies that are currently available at the building, community, and city scales and (2) to develop a planning matrix that evaluates the appropriateness of nutrient management technologies at these three scales based on practical characteristics of import to stakeholders and decision makers. Because each sewershed is unique, the planning matrix developed herein should be customizable to account for priorities of any given municipality, thereby representing a flexible and important tool for sewershed-scale nutrient management. In the current article, we target nitrogen and phosphorus specifically, but our aim is to develop a framework that is flexible enough that it could, in the future, be expanded to also consider organic nutrients or other recoverable resources.

Toward these goals, the planning matrix developed to evaluate the appropriateness of nutrient management technologies is based on the house of quality (HoQ) structure, which is a quality function deployment method typically used in commercial businesses to determine how well a product meets the needs of its customers (Hauser and Clausing, 1988; Park and Kim, 1998). Quality function deployment started in Japan in the 1970s at Mitsubishi and Toyota (Hauser and Clausing, 1988) with the goal of implementing a tool that determines how well a company's product meets the needs of its customers. The HoQ came to the United States in the later 1980s to companies such as Ford, Xerox, General Motors, Campbell's Soup, Colgate, and Fidelity Trust and is used today across several disciplines (Griffin and Hauser, 1993; Ho et al., 2012). However, to the best of our knowledge, the HoQ has not been previously used to evaluate wastewater treatment technologies.

Methods

Planning matrix structure

The structure of the HoQ, as it is applied here, is shown in Fig. 1. The house has five regions, which are elaborated in the following subsections. The left wall, Technologies, is a list of all potential technologies available for nutrient removal or recovery at a given scale. The ceiling, labeled as Technical characteristics, presents the attributes of candidate technologies believed to be important to the client. In the current analysis, we have selected 10 characteristics, but the developer of the HoQ may choose as many characteristics as are relevant to the stakeholders. Importance is the weighting given to each technical characteristic. The central part of the HoQ is the Relationship matrix, where the technologies are evaluated for each of the technical characteristics. The roof of the house, the Correlation matrix, displays the interaction between the technical characteristics (i.e., some of the technical characteristics are likely to be correlated—a technology that scores poorly in material consumption is likely to also score poorly in cost, but other technical characteristics are uncorrelated—how a technology scores in esthetics is not likely to provide any information about how it will score in technical maturity).

FIG. 1.

Modified version of house of quality used to evaluate nutrient removal and recovery technologies at the building, community, and city scales. Adapted from Lowe (2000).

Technologies

A list of nutrient management technologies available at each of the three scales was compiled through a review of literature. As described subsequently, the HoQ enables each technology to be evaluated according to selected technical characteristics and selected weighting for each characteristic.

Technical characteristics

Technologies on the left wall were evaluated by 10 relevant technical characteristics chosen to cover the main attributes that are important for nutrient removal and recovery from wastewater. Operational definitions for each technical characteristic are elaborated in Table 1. The 10 technical characteristics are based on expert opinion and supported by the literature. In addition, HoQ methodology is flexible, and a different user could select different characteristics if those employed here do not account for some other important aspect or consideration.

Table 1.

Abbreviations and General Description for Each of 10 Technical Characteristics

	Technical characteristic	Abbreviation	General description
1	Ease of operation and maintenance	O&M	Lack of complexity and resources needed in technical handling of the technology after it has been installed.
2	Size/footprint	SF	Extent to which the technology requires physical space.
3	Capital cost	CC	Cost to install and ready the technology for use compared with the baseline technology of sewer connections with centralized treatment.
4	Operational cost	OC	Extent to which the product requires operational input in terms of energy, chemicals, water, and labor.
5	Value of end products	EP	Extent to which the technology provides valuable products, such as nutrients, energy, water, and commodities.
6	Technical maturity	TM	Readiness of the technology to be effectively used in nutrient removal or recovery as reported in literature.
7	Environmental impact	EI	Extent of the impact to the environment on the basis of air pollution, land pollution, release of nitrogen, and release of phosphorus.
8	Performance in nitrogen removal	PN	Extent of nitrogen removal or recovery by the technology.
9	Performance in phosphorus removal	PP	Extent of phosphorus removal or recovery by the technology.
10	Esthetics	A	Extent to which the technology is acceptable to stakeholders on the basis of visual appeal, noise level, and odors.

Importance

Weightings for each characteristic, shown in Table 2, were allocated for each of the three scales based on the judgment of the authors. The different weightings for different scales reflect the importance of technical characteristics at a given scale. For instance, ease of operation and maintenance are weighted strongly at the building scale because individual homeowners and building managers will only accept and employ a technology that is easy to use and maintain; operation and maintenance are weighted less strongly at the city scale because a centralized treatment plant has trained professional operators who are dedicated to managing the adopted technologies. At each of the three scales, the average weighting for the 10 characteristics is 1.0.

Table 2.

Weightings for Each Scale for 10 Technical Characteristics

		Weighting
		Building scale	Community scale	City scale
1	O&M	1.5	1.3	0.8
2	SF	1.4	1	0.6
3	CC	1	1	1.1
4	OC	1	1.1	1.1
5	EP	0.5	0.8	1.1
6	TM	1.2	1.1	1.1
7	EI	0.7	1	1.3
8	PN	0.7	0.9	1.2
9	PP	0.7	0.9	1.2
10	A	1.3	0.9	0.5

The importance of one technical characteristic over another for a given assessment can vary greatly depending on the economic and geographical context of the municipality or community needs and priorities; thus, the importance can be modified by the WWTP stakeholders using different weighting. For example, if cost is an overriding consideration for a particular municipality, that municipality could provide a greater weighting for operational cost and concomitantly reduce the weighting for the other characteristics.

Relationship matrix

The relationship matrix reports how well each technology fulfills the technical characteristics chosen by the stakeholder (in this case, the authors). In this article, literature was reviewed for each technology and its associated 10 technical characteristics. The technologies were evaluated according to the rubric for the 10 technical characteristics in Table 3. Technologies that performed well on a given technical characteristic (according to the judgment of the authors) received a + or ++ for that characteristic, while technologies that did not perform well received a − or −−. For example, a review of literature indicates that sidestream chemical precipitation may recover 90% of phosphorus, so according to the rubric in Table 3, it would receive a ++ for performance in phosphorus removal. All of the technical characteristics, except for the capital cost, use an absolute (rather than relative) metric to independently determine the technology's score for that characteristic. The capital cost is scored relative to a baseline scenario of centralized wastewater treatment, which includes flush toilets, combined sanitary and storm sewer systems, gravity sewer, and conventional wastewater treatment. Conventional wastewater treatment is defined as activated sludge, a secondary wastewater treatment with the top priority of carbon removal (and minor nutrient removal in the sludge), but no tertiary treatment of nitrification and denitrification.

Table 3.

Rubric Used to Evaluate Each Nutrient Removal and Recovery Technology Using 10 Technical Characteristics

Technical characteristic	Metric	++	+	0	−	−−
Ease of operation and maintenance	Amount of oversight needed and complexity and input of resources	Minimal oversight needed; very low complexity and input of resources	Little oversight needed; low complexity and input of resources	Some oversight needed; moderate complexity and input of resources	Significant oversight needed; high complexity and input of resources	Very significant oversight needed; very high complexity and input of resources
Size/footprint	Hydraulic retention time of the technology	Seconds	Minutes	Hours	Days	Weeks
Capital cost	Estimated fixed one-time cost to install and ready the technology for use (compared with the baseline scenario)	Estimated to be better than 20% less expensive (for same design capacity) than baseline technology	Estimated to be 10–20% less expensive than baseline technology	Estimated to be within 10% of the baseline cost	Estimated to be 10–20% more expensive than baseline technology	Estimated to be worse than 20% more expensive than baseline technology
Operation and maintenance costs	Number of inputs (energy, chemicals, water, and/or labor) needed to operate and maintain the technology	0 inputs	1	2	3	4
Value of end products	Number and quality of products (nutrients, energy, water, and commodities) recovered by a technology	Creates multiple products or one product of very high quality	Creates two products or one product of moderate quality	Creates one product of marginal products	Creates no valued products	Creates burden product
Technical maturity	Most advanced status reported in literature	Widespread full-scale operation	Limited full-scale operation	Pilot location	Laboratory experiments	No laboratory experiments
Environmental impact	Number of possible adverse effects to the environment caused by the implementation of the technology	0 Pollutants	1 Pollutant	2 Pollutants	3 Pollutants	4 Pollutants
Performance of nitrogen removal	Percentage removal or recovery of nitrogen because of the technology	90+	75–90	50–75	25–50	0–25
Performance of phosphorus removal	Percentage removal or recovery of phosphorus of the technology	90+	75–90	50–75	25–50	0–25
Aesthetics	Number of senses offended (sight, sound, smell) by a technology	Enhances aesthetics	0	1	2	3

Each technology can then be awarded an overall technology score by awarding five points for ++, 4 points for +, etc. The score in each category is multiplied by the importance or weighting factor assigned to that category (listed in Table 2). The overall technology score is the sum of the weighted scores for the 10 technical characteristics as shown by the equation below: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} Overall \ score = & Score_{1} Weighting_{1} + Score_{2} Weighting_{2} \\ & \quad + [\ldots] + Score_{10}Weighting_{10} \end{align*} \end{document}

A higher overall numerical score indicates a higher performing technology. In the Analysis of Treatment Technologies section, the overall numerical score for each technology is compared with the overall numerical score of the baseline scenario mentioned above, allowing us to identify nutrient management alternatives at each scale that might be competitive with, or preferable to, the current technology standard in the context of the priorities of the authors.

Overall numerical scores depend on both the technology's score in each characteristic and the weighting assigned to each characteristic. Both of these can be context specific. Weightings are context specific because different stakeholders or municipalities may have different priorities; in some cities, for example, cost may be an overriding concern, but in others, cost may be secondary to environmental impact or other factors. Furthermore, a technology's score within each characteristic is also context specific; in regions where population density is low, for instance, decentralized or on-site systems (e.g., septic systems) should score better than in regions where population density is high. Because of this dependence on context, the current article attempts to base its analysis on a generic city or municipality, but it must be recognized that the analysis herein is predicated on a set of built-in assumptions, such as the housing density being high enough for conventional centralized treatment to be economically viable.

For each scale, the best two to four technologies with the highest overall numerical scores are discussed in depth in the Analysis of Treatment Technologies section. To select technologies that are best at each scale, we looked for total scores >30 and for a gap of at least 8% between the scores of the best technologies and the other technologies considered.

Correlation matrix

The roof of the HoQ displays reinforcing interactions between technical characteristics with a checkmark and balancing interactions with an X. For example, size/footprint and capital cost would have a checkmark because a larger sized facility (higher size/footprint) would require more money to purchase the property (higher capital cost). In contrast, a more mature technology is more likely to have a reduced capital cost because its history of use could allow for many organizations to design and construct such a technology; this would be indicated by an X at the intersection of operation and maintenance and capital cost.

Analysis of Treatment Technologies

Building-scale house of quality

In developed countries, homes and buildings are equipped with flush toilets that typically use potable water to convey the nutrient-rich waste stream to treatment facilities far outside of the land boundaries within which the building stands. Having treatment processes within the buildings' boundaries provides the opportunity to intercept wastewater where it has the highest nutrient concentrations and lowest volume. More specifically, diverting urine at the point of collection has been studied as a possible solution to reduce nutrient loading to the centralized WWTP (Jimenez et al., 2015). Urine is estimated to contribute 75% of the nitrogen mass load and 50% of the phosphorus mass load to a WWTP, while only contributing 1% of the flow by volume (Larsen and Gujer, 1996).

Table 4 is the HoQ that compares the baseline technology of a conventional toilet (connected to a sewer system and, eventually, a centralized treatment plant) with several other building-scale nutrient management technologies, including struvite precipitation from urine, aerobic and anaerobic membrane bioreactors (MBRs), treatment wetlands, and on-site wastewater treatment systems (OWTS), commonly called septic systems (Crites and Tchobanoglous, 1998). Table 4 includes technologies that treat combined wastewater (graywater and blackwater), blackwater, or diverted urine. Each of these waste streams has a unique composition and requires specific treatment mechanisms that are important considerations (Rashidi et al., 2015).

Table 4.

House of Quality for Building-Scale Nutrient Management Technologies

	Importance			O&M 1.5	SF 1.4	CC 1	OC 1	EP 0.5	TM 1.2	EI 0.7	PN 0.7	PP 0.7	A 1.3
Technical characteristics	Ease of operation and maintenance	O&M	1
	Size/density/footprint	SF	2
	Capital cost	CC	3		✓
	Operational cost	OC	4	✓
	End products	EP	5
	Technical maturity	TM	6	✓		X	X	✓
	Environmental impact	EI	7		✓
	Performance N	PN	8	✓				✓	✓	X
	Performance P	PP	9	✓				✓	✓	X
	Esthetics	A	10		✓	✓		✓
Technologies	Conventional wastewater treatment			++	+	0	+	−−	++	−−	−−	−−	+	33.9
	Composting toilet^a			−	−−	0	+	0	+	++	++	++	−	30.8
	Aerobic MBR			−−	0	−−	0	++	+	0	++	++	0	30.0
	Treatment wetlands			−	−	−−	+	−	+	+	−	−	++	28.7
	Septic systems			+	−	−	0	−−	++	−−	−−	−−	+	27.6
	Direct urine application as fertilizer^b			−−	−−	−−	+	+	0	++	++	++	−	26.6
	Anaerobic MBR			−	0	−−	0	+	0	0	−−	−−	0	24.2
	Nitrification and distillation of urine^b			−−	−−	−−	0	0	−	+	++	++	−	23.2
	Struvite+absorption with zeolites in urine^b			−−	−−	−−	−	+	−	0	+	+	−−	19.3
	Urine ANAMMOX^b			−−	−−	−−	0	0	−	+	++	−−	−−	19.1
	Struvite precipitation from urine^b			−−	−−	−−	−	++	−	+	−−	++	−−	19.1
	NH₃ stripping to H₂SO₄ in urine^b			−−	−−	−−	−	+	−	+	++	−−	−−	18.6
	Anaerobic digestion			−	−−	−−	0	0	+	−			−−	17.4
	Anion exchange in urine^b			−−	−−	−−	−−	0	−	0	−−	++	−−	16.4

Technology able to utilize urine diversion.

Technology dependent on urine diversion.

✓, Positive correlation; MBR, membrane bioreactor; X, negative correlation.

Using the HoQ, a numerical score was calculated for each building-scale technology, as described in the Relationship matrix section. Based on the weightings employed here, conventional wastewater treatment scored the highest (33.9). Two other technologies had scores above 30: composting toilets and aerobic MBRs. Composting toilets, which received a score of 30.8, aerobically treat human waste to create nutrient-rich compost that can be used as a soil amendment in agricultural operations (Anand and Apul, 2014). Composting toilets achieved a relatively high overall score due to theoretically high nutrient recovery, production of a useable product with no pollutants to air, land, or water, and only a slight increase in operational cost. As a result of these characteristics, composting dry toilets have been used in remote locations, such as parks, as a method of managing wastewater. Aerobic MBRs received the third highest score, 30.0, as shown in Table 4. At the building scale, aerobic MBRs can treat wastewater to reclaimed water quality. These systems have been placed inside green buildings for the ability to produce reclaimed water by removing nutrients on a relatively small footprint (Rashidi et al., 2015). This technology received a ++ in the following categories: end products, performance in N removal, and performance in P removal.

The top score of conventional wastewater treatment highlights the difficulty of bringing nutrient recovery technologies into the building scale when conventional wastewater treatment is the status quo. Conventional treatment is easy for the user to maintain (i.e., just a toilet and home plumbing connected to a sewer main), has a small footprint, and is not offensive esthetically. The technical characteristics with the highest weighting at the building scale are operation and maintenance (1.5), size/footprint (1.4), and esthetics (1.3); only conventional wastewater treatment scored positively across all three of these technical characteristics. The technologies that scored most closely to the baseline technology are those that have high performance in removal or recovery of nitrogen, have high performance in removal or recovery of phosphorus, and have low impacts on the environment. The composting toilet had the second highest score despite having negative scores in operation and maintenance, size/footprint, and esthetics. Technologies that did not perform well in this evaluation not only had negative scores in operation and maintenance, size/footprint, and esthetics but also in technical maturity and several other technical characteristics.

Septic systems, which are deployed widely in cases where conventional treatment is not viable (e.g., low population density), received a lower score (27.6) than other building-scale technologies such as compositing toilets (30.8), aerobic MBRs (30.0), and treatment wetlands (28.7). Septic systems have lower scores than the three alternative technologies for the technical characteristics of end products, environmental impact, nitrogen performance, and phosphorus performance. Therefore, in cases where a high degree of decentralization is required, these three technologies may be preferable.

Community-scale house of quality

Community-scale technologies have the benefit of treating wastewater in moderate volume while often being closer to the waste source (and/or potential reuse location) than a centralized treatment plant. Community-scale technologies predominantly incorporate technologies that require minimal maintenance and oversight, incur low operational costs, and are technically mature. Most of the technologies listed in the community-scale HoQ shown in Table 5 use physical and biological treatment processes to remove contaminants (Onsite Wastewater Treatment Systems Manual, 2002; Massoud et al., 2009; Makropoulos and Butler, 2010). An activated sludge system was used as the baseline technology to compare the capital costs.

Table 5.

House of Quality for Community-Scale Nutrient Management Technologies

	Importance		O&M 1.3	SF 1	CC 1	OC 1.1	EP 0.8	TM 1.1	EI 1	PN 0.9	PP 0.9	A 0.9
Technical characteristics	Ease of operation and maintenance	O&M
	Size/footprint	SF
	Capital cost	CC		✓
	Operational cost	OC	✓
	Value of end products	EP
	Technical maturity	TM	✓		X	X	✓
	Environmental impact	EI		✓
	Performance (N)	PN	✓				✓	✓	X
	Performance (P)	PP	✓				✓	✓	X
	Esthetics	A		✓	✓		✓
Technologies	Constructed wetlands		++	−−	++	++	−	++	0	0	0	++	38.0
	Facultative lagoons		++	−−	+	+	−−	++	0	+	−	−	32.4
	Rotating biological contactor		−	0	−	−	+	++	0	++	+	0	32.3
	BNR		−−	0	−	−	+	++	0	++	++	0	31.9
	Algal photobioreactors		−	−	−	−	+	0	+	++	++	0	31.0
	Sequencing batch reactors—BNR mode		−−	0	−	−	+	+	0	++	++	0	30.8
	Communal septic systems		++	−	+	++	−−	++	−	−−	−−	0	30.8
	Recirculating Sand filter		0	−	0	+	−−	++	0	0	−	+	30.7
	Intermittent sand filter		+	−	0	+	−−	++	0	−	−	0	30.2
	MBRs—BNR mode		−−	0	−−	−−	++	+	0	++	++	0	29.5
	Upflow anaerobic sludge blanket		−	0	0	++	+	+	0	−−	−−	0	29.2
	Aerated lagoons		+	−−	0	+	−−	++	0	0	−−	0	29.2
	Activated sludge systems		−	0	0	0	0	++	0	−	−	0	29.1
	Anaerobic lagoons		++	−−	+	++	−−	++	−	−−	−−	−	28.9
	Algal MBRs		−	−	−−	−−	++	−	+	++	++	0	28.6
	Anaerobic MBRs		−	0	−−	+	+	0	0	−−	−−	0	25.0

BNR, biological nutrient removal.

Using the HoQ, a numerical score was calculated for each community-scale technology, as described in the Relationship matrix section. Based on the weightings employed here, constructed wetlands (38.0) scored the highest, followed by facultative lagoons (32.4), rotating biological contactors (RBCs) (32.3), and biological nutrient removal (BNR) (31.9). Constructed wetlands are artificially engineered wetlands treating wastewater through processes involving uptake by vegetation, soil absorption, sedimentation, and microbial activity. The preference toward constructed wetlands is primarily due to its high scores in operation and maintenance, operational costs, technical maturity, and esthetics. Facultative lagoons utilize layers with different dissolved oxygen levels to treat wastewater without mechanical mixing or aeration. Facultative lagoons are not as esthetically pleasing as constructed wetlands. BNR, defined here as the removal of N and P using a combination of nitrification, denitrification, and enhanced biological phosphorus removal processes, may be implemented on its own or in combination with different types of reactor systems such as MBRs. BNR technologies create moderate- to high-quality reclaimed water. BNR and MBRs have high scores in end products, technical maturity, nitrogen performance, and phosphorus performance.

Constructed wetlands (which achieved the top score by several points) and facultative lagoons separated themselves from the competition based on cost (high scores in operation and maintenance and operational costs). Accordingly, these two technical characteristics were deemed to have the most importance at the community scale (along with technical maturity, which showed significantly less variation in scores). However, neither constructed wetlands, nor facultative lagoons, nor RBCs easily facilitate recovery of N or P in a readily usable form. BNR was competitive because of its high scores in end products, nitrogen performance, and phosphorus performance. If the future brings increased demand for nutrient recovery, along with technical advances in making community-scale technologies easy and cost-effective to operate and maintain, it may be preferable to install RBC, BNR, or MBR technologies rather than the low-tech, but cost-effective, wetlands or lagoons.

City scale: mainstream house of quality

City-scale technologies for mainstream wastewater treatment are common in urban settings due to their ability to treat large volumes of water in one central location. To protect the health of ecosystems that receive treatment plant discharge, mainstream technologies reduce the environmental impact of collected sewage by removing carbon, nitrogen, and phosphorus. Several mainstream technologies are evaluated in Table 6 using the HoQ. To evaluate capital costs, technologies are compared against a baseline scenario of activated sludge for carbon removal, separate-stage nitrification/denitrification for nitrogen removal, and alum addition for phosphorus removal; technologies in Table 6 that remove only one element (C, N, or P) are compared against the relevant treatment process that removes that element.

Table 6.

House of Quality for City-Scale Mainstream Nutrient Management Technologies

	Importance			O&M 0.8	SF 0.6	CC 1.1	OC 1.1	EP 1.1	TM 1.1	EI 1.3	PN 1.2	PP 1.2	A 0.5
Parameters	Ease of operation and maintenance	O&M	1
	Size/footprint	SF	2
	Capital cost	CC	3		✓
	Operational cost	OC	4	✓
	Value of end products	EP	5
	Technical maturity	TM	6	✓		X	X	✓
	Environmental impact	EI	7		✓
	Performance N	PN	8	✓				✓	✓	X
	Performance P	PP	9	✓				✓	✓	X
	Esthetics	A	10		✓	✓		✓
Technologies	A²O			++	0	−	0	−	++	+	++	++	0	37.7
	Oxidation ditch			−	0	+	0	−	++	+	++	++	0	37.5
	Five-stage Bardenpho			+	0	−	0	−	++	+	++	++	0	36.9
	Membrane technologies			−	0	−−	−	−	++	+	++	++	0	33.1
	Chemically enhanced primary treatment			+	+	++	−	−	+	0	+	−	0	32.5
	Fixed film nitrification/denitrification			−	0	+	0	−	++	0	++	−−	0	31.4
	Conventional separate stage Nitrification/denitrification			0	0	0	−	−	++	+	++	−−	0	31.3
	Shortcut nitrogen removal			+	0	0	−	−	++	+	++	−−	0	30.6
	Chemical precipitation and crystallization			0	0	0	−	−	++	0	−−	++	0	30.0
	Mechanical separation			+	+	++	0	−	++	−	−−	−−	0	28.6
	Enhanced biological phosphorus removal			−	0	0	−	−	++	0	−−	+	0	28.0
	ANAMMOX			−	0	−	−	−	0	+	+	−−	0	26.0
	Sludge ash recovery			−	0	−	−	++	−	0	−−	0	0	25.7

A²O, anaerobic/anoxic/oxic.

A large number of mainstream technologies are available, but many of these technologies are similar in principle and operation, varying mainly in configurational details; therefore, in Table 6, similar technologies are grouped together and not all candidate technologies are included. For example, anaerobic/anoxic/oxic (A²O) treatment is included, but anoxic/oxic (A/O) and modified Ludzack-Ettinger (MLE) processes are not included because A²O can be considered a combination of A/O and MLE. Similarly, fixed-film nitrification–denitrification includes both trickling filters and RBCs.

Based on the weightings employed here, the three highest scoring technologies were A²O (37.7 points), oxidation ditch (37.5 points), and five-stage Bardenpho (36.9 points). The A²O process achieves high removal of both nitrogen and phosphorus by placing an anaerobic chamber before the anoxic and aerobic chambers. It receives high scores for O&M, TM, EI, PN, and PP. The oxidation ditch receives influent in its anaerobic reactor, which is followed by alternating anoxic and aerobic zones. The oxidation ditch scored highly in CC, TM, EI, PN, and PP. The five-stage Bardenpho utilizes additional anoxic and aerobic reactors to meet requirements of low TN and low TP and has similar scores to A²O.

In contrast to the baseline technology of separate stage nitrification–denitrification, all three of the highest scoring technologies make use of carbon that is already present in wastewater to drive denitrification, thereby saving money on operation and maintenance by not needing an external carbon source. Candidate technologies for city-scale mainstream treatment that did not perform well in this evaluation are those that do not recover valuable resources, lack maturity, and/or require significant amounts of money, chemicals, or energy to operate and maintain.

City scale: sidestream house of quality

During city-scale mainstream treatment, anaerobic digestion is often employed to treat the sludge from primary and secondary treatments. Effluent of anaerobic digestion includes biogas, biosolids, and a liquid effluent stream that is typically recycled back to the beginning of the mainstream treatment process. This liquid effluent stream is often called the sidestream. Compared with the mainstream, the sidestream contains higher concentrations of nitrogen and phosphorus, lower flow rates, and lower levels of carbon. Therefore, mainstream technologies that rely on higher COD:N ratios for denitrification, such as A²O, Bardenpho, and oxidation ditch, are not considered for sidestream treatment. A baseline scenario for sidestream treatment is return to the headworks without additional treatment; treatment of the sidestream is presently increasing in popularity, but is not yet commonly used.

Several sidestream technologies are evaluated in Table 7 using the HoQ. The two top-scoring technologies were ion exchange and chemical precipitation and crystallization. Ion exchange, which had a score of 36.1, recovers nutrients that can be used as fertilizer in the sidestream, a more favorable location for recovery (compared with mainstream treatment) due to higher concentrations of nitrogen and phosphorus. One example of an ion exchange technology is the RIM-NUT process, which uses the natural zeolite clinoptilolite for ion exchange and a strong base resin for regeneration and subsequent reuse (Liberti et al., 1987). The process recovers 90% of nitrogen, which can be used for fertilizer. Ion exchange can be integrated with chemical precipitation to recover both nitrogen and phosphorus. Negative scores are given for operation and maintenance, capital cost, and operational cost, mostly due to the cost of resin and materials; ++ scores were given for end products, environmental impact, nitrogen performance, and phosphorus performance. Recovering N and P in the sidestream offers the dual benefits of producing a potentially valuable product and reducing the cost of removing N and P during mainstream treatment (which would otherwise be required if a nutrient-rich sidestream is returned to the plant headworks).

Table 7.

House of Quality for City-Scale Sidestream Nutrient Management Technologies

	Importance		O&M 0.8	SF 0.6	CC 1.1	OC 1.1	EP 1.1	TM 1.1	EI 1.3	PN 1.2	PP 1.2	A 0.5
Technical characteristics	Ease of operation and maintenance	O&M
	Size/footprint	SF
	Capital cost	CC		✓
	Operational cost	OC
	Value of end products	VEP
	Technical maturity	TM	✓		X	X	✓
	Environmental impact	EI		✓
	Performance (N)	PN	✓				✓	✓	X
	Performance (P)	PP	✓				✓	✓	X
	Esthetics	A			✓
Technologies	Ion exchange		−	+	−−	−	++	0	++	++	++	0	36.1
	Chemical precipitation and crystallization		0	+	−	−	++	++	+	−	++	0	35.3
	Ammonia stripping		0	0	−	−	++	+	+	++	−−	0	32.4
	No treatment		−	++	0	++	−	++	++	−−	−−	+	32.0
	Forward osmosis		0	0	−−	−	−	−	++	++	++	0	31.9
	Conventional nitrification–denitrification		0	0	−	−	−	++	+	++	−−	0	30.2
	Microbial fuel cell		0	0	−−	0	0	0	+	++	−−	0	29.1
	Shortcut nitrogen removal/recovery (ANAMMOX, SHARON, CANON, SHARON-ANAMMOX, DEAMOX)		−	0	−	−	−	++	+	+	−−	0	28.2
	Breakpoint chlorination		−	−	−	−	−−	+	+	+	−−	0	25.4

Chemical precipitation and crystallization, which use the addition of a divalent or trivalent metal salt to remove phosphorus (and, to a lesser extent, nitrogen) through sedimentation of the precipitate (Jenkins et al., 1971; Halling-Sorensen and Jorgensen, 1993), received a score of 35.3. Cations such as calcium, iron, and aluminum can be added to bind with phosphate within a fluidized reactor to be settled and recovered. However, the addition of Fe³⁺ or Al³⁺ may not produce a recoverable product that has a market to be sold. If Mg²⁺ is added, as in the case of struvite precipitation, a fertilizer product (MgNH₄PO₄) can be recovered. Struvite precipitation typically recovers 80–95% of the phosphorus in the sidestream; however, only 10–40% of the nitrogen is typically recovered during the process (Water Environment Research Foundation [WERF], 2012). Using the example of struvite precipitation, this technology receives ++ scores in end products and technical maturity due to recovering the engineered struvite precipitate and subsequent reduction in likelihood of nuisance struvite precipitation. However, it receives negative scores in capital cost, operational cost, and nitrogen performance primarily because of chemical addition. Companies that offer a process to produce a struvite product include Aquatec Maxcon (Crystalactor®) and Ostara. Engineered struvite precipitation reduces the potential for nuisance struvite precipitation and, as with ion exchange, also reduces the amount of nutrients needing treatment in the mainstream.

Overall, sidestream technologies such as ion exchange and chemical precipitation and crystallization scored the highest because of their ability to remove nitrogen while also recovering valuable end products that can be used as a fertilizer.

Discussion

Expected future trends

Rankings at each scale generally align with current wastewater treatment practice. For instance, at the building scale, conventional treatment is ranked highest because of its easy maintenance, small footprint, and inoffensive esthetics. Similarly, at the city scale, top-ranked technologies are those that are commonly employed (e.g., A²O, oxidation ditch) and use the dissolved organic carbon present in wastewater to drive denitrification.

However, future trends will likely affect the technologies, weightings, and scores and therefore change the ranking of the technologies. One trend is diminishing phosphorus reserves, which are likely to be depleted in the next 50–100 years (Cordell et al., 2009). While the supply is diminishing, phosphorus demand is expected to increase until it reaches its peak demand around 2030 (Cordell et al., 2009). This trend is exacerbated by the uneven global distribution of phosphorus reserves. Thus, the value of end products would receive higher weightings in the future as P recovery becomes more important and provides more revenue. Another trend is continued research and development of wastewater treatment technologies, which may result in higher scores across all 10 technical characteristics for up-and-coming technologies. Additionally, implementation of test beds can provide opportunities to improve the technical characteristics of developing technologies while minimizing risk for municipalities (Water Environment Research Foundation [WERF], 2017). The trend of continued research and development of wastewater treatment technologies is especially noticeable at the building and community scales, where several publications highlight the need for and development of source separation and decentralization technologies (Larsen et al., 2013; Skambraks et al., 2017). As these technologies develop and become easier to operate and maintain, reduce in size, and improve esthetically, they will challenge the current paradigm of a flush toilet connected to a septic tank or a sewer system. Building- and community-scale technologies, because of their more decentralized nature, are more nimble and can produce cost savings due to reduced idle capacity when population growth is less than predicted (Roefs et al., 2017). A path forward is distributed systems, which combine both centralized and decentralized treatment technologies.

Limitations of this analysis

One limitation of this article is that the weightings cannot be universally applied to every community. Scores were assigned to each technology and weightings applied to each technical characteristic without any specific community in mind; thus, the analysis is somewhat generic. Therefore, each city's particular context might override the scores and weightings assigned based on its unique geography, financial resources, population, and preferences. For example, some centralized plants may be located in areas where residents are highly concerned about esthetics and would not choose any technology that creates odors. However, our easily customizable planning matrix can accommodate a situation such as this because each community can adjust the weightings of the ten characteristics (e.g., esthetics) to determine the most appropriate city-, community-, and building-scale nutrient management technologies.

Another limitation of this article is that the ramifications of new upstream treatment on existing downstream treatment are unknown. The introduction of building-scale and community-scale technologies could produce a number of consequences, such as reduced nutrient loading on existing city-scale WWTPs or reduced costs of centralized treatment. However, the reduced nutrient loading could prevent a centralized plant from economically precipitating and selling struvite. Therefore, future research is needed that evaluates how the introduction of new upstream nutrient management technologies affects treatment efficiency and economics of nutrient management across an entire sewershed.

Relationship to other decision-making tools

The HoQ planning matrix described herein is an efficient tool because Tables 4 –7—which list all technologies, their grades across 10 technical characteristics, and a default importance factor—can be easily adjusted to accommodate different contexts. This method is complementary and can be used in conjunction with, or instead of, other methods such as the eco-balance (e.g., Kimura and Hatano, 2007), life cycle assessment (e.g., Cornejo et al., 2016), stakeholder analysis (e.g., Lienert et al., 2013), and multicriteria decision analysis (e.g., Flores-Alsina et al., 2008). Eco-balance, which identifies the environmental impact of various business activities, and life cycle assessment, which identifies and analyzes the environmental impact of various processes or products, provide depth on one technical characteristic such as environmental impact, but do not provide the breadth of ten technical characteristics. Stakeholder analysis identifies, prioritizes, and understands key stakeholders, but does not provide decision-making support. The HoQ utilizes elements of multicriteria analysis such as defining the decision, identifying stakeholder interests, weighting stakeholder interests, and scoring alternatives. The planning matrix developed here is unique in that it utilizes ten technical characteristics based on expert opinion, is supported by literature, and evaluates wastewater treatment technologies based on their scale of application.

Footnotes

Acknowledgments

This publication was made possible by the U.S. Environmental Protection Agency (USEPA) grant 8556901. Its contents are solely the responsibility of the grantee and do not necessarily represent the official views of the USEPA. Furthermore, USEPA does not endorse the purchase of any commercial products or services mentioned in the publication.

Author Disclosure Statement

No competing financial interests exist.

References

Anand

C.K.

, and Apul

D.S.

(2014). Composting toilets as a sustainable alternative to urban sanitation—A review. Waste Manag. 34, 329.

Cai

, Park

S.Y.

, and Li

(2013). Nutrient recovery from wastewater streams by microalgae: Status and prospects. Renew. Sust. Energ. Rev., 19, 360.

Cordell

, Drangert

, and White

(2009). The story of phosphorus: Global food security and food for thought. Global Environ. Chang., 19, 292.

Cornejo

P.K.

, Zhang

, and Mihelcic

J.R.

(2016). How does scale of implementation impact the environmental sustainability of wastewater treatment integrated with resource recovery? Environ. Sci. Technol., 50, 6680.

Crites

R.W.

, and Tchobanoglous

(1998). Small and Decentralized Wastewater Management Systems. Boston: WCB/McGraw-Hill.

De-Bashan

L.E.

, and Bashan

(2004). Recent advances in removing phosphorus from wastewater and its future use as fertilizer (1997–2003). Water Res. 38, 4222.

Doyle

J.D.

, and Parsons

S.A.

(2002). Struvite formation, control and recovery. Water Res. 36, 3925.

Englehardt

J.D.

, Wu

, Bloetscher

, Deng

, du Pisani

, Eilert

, Elmir

, Guo

, Jacangelo

, LeChevallier

, Leverenz

, Mancha

, Plater-Zyberk

, Sheikh

, Steinle-Darling

, and Tchobanoglous

(2016). Net-zero water management: Achieving energy-positive municipal water supply. Environ. Sci. Water Res. Tech., 2, 250.

Flores-Alsina

, Rodriguez-Roda

, Sin

, and Gernaey

K.V.

(2008). Multi-criteria evaluation of wastewater treatment plant control strategies under uncertainty. Water Res. 42, 4485.

10.

Gikas

, and Tchobanoglous

(2009). The role of satellite and decentralized strategies in water resources management. Environ. Manage., 90, 144.

11.

Griffin

, and Hauser

J.R.

(1993). The voice of the customer. Market. Sci., 12, 1.

12.

Guest

J.S.

, Skerlos

S.J.

, Barnard

J.L.

, Beck

M.B.

, Daigger

G.T.

, Hilger

, Jackson

S.J.

, Karvazy

, Kelly

, Macpherson

, Mihelcic

J.R.

, Pramanik

, Raskin

, van Loosdrecht

M.C.M.

, Yeh

, and Love

N.G.

(2009). A new planning and design paradigm to achieve sustainable resource recovery from wastewater. Environ. Sci. Technol., 43, 6126.

13.

Halling-Sorensen

, and Jorgensen

S.E.

(1993). The Removal of Nitrogen Compounds from Wastewater. Amsterdam: Elsevier.

14.

Hauser

, and Clausing

(1988). The house of quality. Harv. Bus. Rev., 66, 63.

15.

Heidler

, Sapkota

, and Halden

R.U.

(2006). Partitioning, persistence, and accumulation in digested sludge of the topical antiseptic triclocarban during wastewater treatment. Environ. Sci. Technol., 40, 3634.

16.

, He

, Lee

C.K.M.

, and Emrouznejad

(2012). Strategic logistics outsourcing: An integrated QFD and fuzzy AHP approach. Expert. Syst. Appl., 39, 10841.

17.

Jenkins

, Ferguson

J.F.

, and Menar

A.B.

(1971). Chemical processes for phosphate removal. Water Res. 5, 369.

18.

Jimenez

, Bott

, Love

, and Bratby

(2015). Source separation of urine as an alternative solution to nutrient management in biological nutrient removal treatment plants. Water Environ. Res., 87, 2120.

19.

Kimura

S.D.

, and Hatano

(2007). An eco-balance approach to the evaluation of historical changes in nitrogen loads at a regional scale. Agric. Syst., 94, 165.

20.

Larsen

T.A.

, and Gujer

(1996). Separate management of anthropogenic nutrient solutions (human urine). Water Sci. Technol., 34, 87.

21.

Larsen

T.A.

, Udert

K.M.

, and Lienert

(2013). Source Separation and Decentralization for Wastewater Management. London: IWA Publishing.

22.

Liberti

, Laricchiuta

, Lopez

, Passino

, and Boari

(1987). The RIM-NUT process at west bari for removal of nutrients from wastewater: Second demonstration. Resour. Conserv., 15, 95.

23.

Lienert

, Schnetzer

, and Ingold

(2013). Stakeholder analysis combined with social network analysis provides fine-grained insights into water infrastructure planning processes. Environ. Manage., 125, 134.

24.

Lowe

A.J.

(2000). House of Quality. Available at: www.webducate.net/qfd/qfd.html (accessed May 17, 2017).

25.

Lundin

, Bengtsson

, and Molander

(2000). Life cycle assessment of wastewater systems: Influence of system boundaries and scale on calculated environmental loads. Environ. Sci. Technol., 34, 180.

26.

Makropoulos

C.K.

, and Butler

(2010). Distributed water infrastructure for sustainable communities. Water Resour. Manag., 24, 2795.

27.

Massoud

M.A.

, Tarhini

, and Nasr

J.A.

(2009). Decentralized approaches to wastewater treatment and management: Applicability in developing countries. Environ. Manage., 90, 652.

28.

, and Zhang

(2013). Energy-nutrients-water nexus: Integrated resource recovery in municipal wastewater treatment plants. Environ. Manage., 127, 255.

29.

Morse

G.K.

, Brett

S.W.

, Guy

J.A.

, and Lester

J.N.

(1998). Review: Phosphorus removal and recovery technologies. Sci. Total Environ., 212, 69.

30.

Onsite Wastewater Treatment Systems Manual. (2002). Washington, DC: US Environmental Protection Agency. Available at: www.epa.gov/septic/onsite-wastewater-treatment-and-disposal-systems (accessed May 17, 2017).

31.

Park

, and Kim

(1998). Determination of an optimal set of design requirements using house of quality. J. Oper. Manag., 16, 569.

32.

Rashidi

, Hoseini

A.G.

, Hoseini

A.G.

, Sluaiman

N.M.N.

, Tookey

, and Hashim

N.A.

(2015). Application of wastewater treatment in sustainable design of green built environments: A review. Renew. Sust. Energ. Rev., 49, 845.

33.

Roefs

, Meulman

, Vreeburg

J.H.G.

, and Spiller

(2017). Centralised, decentralized or hybrid sanitation systems? Economic evaluation under urban development uncertainty and phased expansion. Water Res. 109, 274.

34.

Skambraks

, Kjerstadius

, Meier

, Davidsson

, Wuttke

, and Giese

(2017). Source separation sewage systems as a trend in urban wastewater management: Drivers for the implementation of pilot areas in Northern Europe. Sust. Cities Society, 48, 287.

35.

Tchobanoglous

, Burton

F.L.

, and Stensel

H.D.

(2004). Wastewater Engineering: Treatment and Reuse. New York: McGraw-Hill.

36.

Teerlink

, Hering

A.S.

, Higgins

C.P.

, and Drewes

J.E.

(2012). Variability of trace organic chemical concentrations in raw wastewater at three distinct sewershed scales. Water Res. 46, 3261.

37.

The Johnson Foundation at Wingspread. (2014). Optimizing the Structure and Scale of Urban Water Infrastructure: Integrating Distributed Systems. Racine: The Johnson Foundation at Wingspread.

38.

Verstraete

, de Caveye

P.V.

, and Diamantis

(2009). Maximum use of resources present in domestic “used water.”. Bioresour. Technol., 100, 5537.

39.

Water Environment Federation (WEF). (2014). Moving Toward Resource Recovery Facilities. Washington, DC: Water Environment Federation.

40.

Water Environment Research Foundation (WERF). (2011). When to Consider Distributed Systems in an Urban and Suburban Context. Available at: www.werf.org/distributedwater (accessed May 17, 2017).

41.

Water Environment Research Foundation (WERF). (2012). Nutrient Recovery: A Key Component of Sustainable Nutrient Removal. Washington, DC. Available at: http://klasystems.com/news/wp-content/uploads/2016/05/Track_01_1100_Nutrient_Recovery.pdf (accessed May 17, 2017).

42.

Water Environment Research Foundation (WERF). (2017). LIFT Water Resource Recovery Test Bed Facility Network. Available at: www.werf.org/lift/Test_Bed_Network/lift/docs/The_LIFT_Water_Resource_Recovery_Test_Bed_Facility_Network.aspx (accessed May 17, 2017).

43.

Yeoman

, Stephenson

, Lester

J.N.

, and Perry

(1988). The removal of phosphorus during wastewater treatment: A review. Environ. Pollut., 49, 183.