Journal of Cleaner Production 262 (2020) 121435

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Evaluating the environmental and economic impact of fruit and

vegetable waste valorisation: The lettuce waste study-case
Stella Plazzotta a, *, Mattia Cottes b, Patrizia Simeoni b, Lara Manzocco a
a
Department of Agricultural, Food, Environmental and Animal Sciences, University of Udine, Italy
b
Polytechnic Department of Engineering and Architecture, University of Udine, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The fruit and vegetable sector generates large amounts of waste, which poses both environmental and
Received 25 September 2019 economic issues. Different strategies can be applied to valorise fruit and vegetable waste (FVW) by
Received in revised form turning it into value-added products. However, the economic and environmental impact of such stra-
31 January 2020
tegies is largely unknown. In this paper, the environmental and economic impact of FVW valorisation on
Accepted 29 March 2020
Available online 2 April 2020
an industrial scale was evaluated by developing a Decision Support System (DSS). To this aim, the lettuce
waste study-case was considered, since different innovative laboratory-scale strategies have been
Handling editor: Dr. Govindan Kannan recently proposed for its valorisation. Investment and running costs, energetic demand and yields of
lettuce waste valorisation processes were collected based on laboratory tests and industrial surveys. The
Keywords: application of the DSS estimated that if 30% of lettuce waste annually produced by a large company was
Food waste valorisation valorised by using a system configuration that involves not only anaerobic digestion and composting, but
Feasibility also high pressure homogenisation to produce fresh juices, and ultrasound-assisted extraction to pro-
Sustainability duce antioxidant extracts, this configuration would lead to an investment lower than 10 million V, a 1
Multi criteria decision making
year-pay-back time and a 72 tons-reduction of carbon dioxide emissions, thus representing a rational
compromise between economic returns and environmental advantage.
The developed multi-objective DSS is a valuable tool to identify the most sustainable and investment-
worthy processes for the valorisation of FVW.
© 2020 Elsevier Ltd. All rights reserved.

1. Introduction dedicated to the development of policies for FVW management

(2006/12/EC). In particular, fruit and vegetable waste (FVW) val-
According to the estimation of the United Nations Food and orisation into value-added derivatives has been extensively and
Agriculture Organization (FAO), every year, one-third of the food increasingly studied in the last years, as evidenced by the enormous
produced in the world (about 1.3 billion metric tons) is wasted number of relevant publications (Supplementary Fig. S1). Despite
(FAO, 2014). The waste generated by the fruit and vegetable sector this intense research activity, the current destination of FVW is
(fruit and vegetable waste, FVW) is estimated to represent up to mainly represented by landfilling, composting, anaerobic digestion
60% of the food waste generated annually in the world (Gustavsson and carbonisation (Cristo  bal et al., 2018). However, when FVW is
et al., 2011). In this regard, roughly one-third of all fruits and veg- used this way, as a feedstock to produce energy and fertilizers, its
etables produced worldwide are wasted before reaching the con- interesting functional molecules are underutilised or lost
sumer (Kader, 2005), and this estimation was reported to be even (Pfaltzgraff et al., 2013). The latter are instead maximally exploited
higher in Italy (about 87%) (Segre
and Falasconi, 2011). FVW rep- when FVW serves as a source of bioactive compounds, functional
resents a significant economic issue for companies and poses food ingredients and biocompatible materials (Papargyropoulou
environmental problems, due to its high moisture and biodegrad- et al., 2014).
ability. Moreover, it involves a loss of valuable biomass and nutri- It must be noted that the valorisation of FVW is at an early stage
ents. For these reasons, in the last decade, large effort has been of development and that essential elements must be still clarified to
assess its viability (Cristo bal et al., 2018; Heck and Rogers, 2014).
Firstly, data on the exact amount of waste produced from food
* Corresponding author. processing is nowadays very limited. In fact, no official data are
E-mail address: stella.plazzotta@uniud.it (S. Plazzotta). issued annually by the EU, given the commercial sensitivity of this

https://doi.org/10.1016/j.jclepro.2020.121435
0959-6526/© 2020 Elsevier Ltd. All rights reserved.
2 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435

information for the producers. Moreover, published data are often using high pressure homogenisation, ultrasounds, air-drying and
largely variable, due to methodological variations and to the supercritical-CO2-drying, respectively (Plazzotta and Manzocco,
intrinsic compositional variability of fruit and vegetables 2018; Plazzotta and Manzocco, 2019; Plazzotta et al., 2018a,
(Pfaltzgraff et al., 2013). Moreover, the demand of resources (e.g. 2018b, 2018c). To this purpose, the paper was structured as fol-
energy, water, workforce, investment for equipment) of valor- lows: chapter 2 describes the DSS (paragraph 2.1), the mathemat-
isation strategies as compared to that of traditional waste man- ical model for lettuce waste management system (paragraph 2.2),
agement options should be evaluated. In fact, the implementation and the simulation system design (paragraph 2.3); in chapter 3 the
of innovative valorisation strategies is viable only if bringing obtained results are reported and discussed; finally, in chapter 4 the
environmental (e.g. reduction of carbon emission and of waste main conclusions that can be drawn from the study are
amount) and economic advantages (e.g. reduction of waste man- summarized.
agement costs and increase of revenues from sale of FVW valor-
isation products) as compared to traditional management 2. Materials and methods
strategies. It is noteworthy that most of literature studies dealing
with FVW valorisation generally do not discuss these crucial 2.1. Decision support system
environmental and economic aspects, limiting the investigation to
the optimization of the process used to turn the waste material into A classical DSS model was applied to lettuce waste valorisation.
a value-added derivative (Galanakis, 2013). However, many of these Its structure consisted of three major phases, whose description
studies exploit innovative technologies such as ultrasounds, mi- and main outputs are described in Fig. 1: investigative phase,
crowaves, high pressure and supercritical fluid processing, which design phase and scenario analysis (Mattiussi et al., 2014; Simeoni
are well-known to require huge investment and maintenance costs, et al., 2018). Initially, the investigative phase aims at collecting
as well as specialized know-how and plants (Galanakis, 2013; quantitative data on the considered industrial system. Subse-
Meullemiestre et al., 2016; Sicaire et al., 2016; Talens et al., 2016). quently, in the design phase, input and output variables, their in-
Also, even when using technologies characterised by a high in- teractions, system constraints and objectives are defined, and
dustrial maturity level (e.g. air-drying) (Alongi et al., 2019; Ferreira combined in multiple scenarios. Finally, in the scenario analysis
et al., 2015) an accurate cost-benefit analysis should be performed phase, the latter are scheduled and compared based on the study
to evaluate the environmental and economic sustainability of the objectives (Simeoni et al., 2018).
proposed FVW valorisation strategies (Meullemiestre et al., 2016; For the investigative phase, the tools of Material Flow Analysis
Sicaire et al., 2016). (MFA) and techno-economic and profitability assessment were
Finally, according to the most recent studies, the feasibility used. All collected data and their analysis referred to a waste
assessment of FVW valorisation cannot be adequately estimated treatment plant following typical annual working period in Italy,
without considering the vast potential of integrated technologies corresponding to 8 daily working hours for 200 annual working
for FWV valorisation in achieving sustainable and carbon-efficient days. The design phase is the core of the used model and is
biorefineries (Yu et al., 2019). In fact, the same waste material can composed by three subsequent stages (Fig. 1):
be subjected to multiple valorisation options by means of inte-
grated biorefinery interventions aiming at reducing waste to zero. - Design of experiment (DOE). DOE was used to classify the sys-
The waste proportion to be destinated to each of these options is to tem variables and to define the system constraints. In particular,
be selected after evaluation of a large number of technical (e.g. as reported in paragraph 2.2, the following quantities were set
integrability of different processes, process scalability), economic as input variables: the initial amount of lettuce waste available
(e.g. investment cost, market price of the valorisation output) and for valorisation; the partition of lettuce waste into traditional
environmental (e.g. process requirement of solvents, water and waste management options or valorisation processes; the price
energy) variables (Cristo bal et al., 2018; Goula and Lazarides, 2015; of valorisation outputs; the value of energy saving incentives.
Lam et al., 2018; Xiong et al., 2019; Yu et al., 2019). The environmental advantage and economic effort indexes
In view of the above, the choice of a FVW valorisation system is a identified in the investigative phase were set as output vari-
multi-criteria decision making (MCDM) process and, like all multi- ables: saved energy, greenhouse gas reduction, total investment
criteria evaluation problems, it faces the challenge of how to cost and payback time.
determine the preferred outcomes given the presence of more than The DOE constraints were based on technical and economic is-
one assessment criterion. The models supporting MCDM are sues. In particular, at least 50% of total lettuce waste was allo-
designed to provide a framework for assessing this information on cated to traditional management strategies, which represent an
preferences in combination with deterministic or empirical infor- important source of biogas and fertilizers. Moreover, selected
mation, so that decisions involving the assessment of multiple lettuce waste, deriving from removal of spoiled and bruised
criteria can be reached within a structured framework. In this re- parts and washing of waste, was set at a value lower than 50% of
gard, the multi-objective method described by Simeoni et al. (2018) the initial lettuce waste weight intended for innovative valor-
could represent a valuable tool to estimate the environmental and isation, due to the probable poor conditions of waste. In addi-
economic implications related to the integration of FVW valor- tion, a payback time higher than 5 years was not considered,
isation strategies in the traditional waste management system, on since not economically advantageous (Heck and Rogers, 2014).
an industrial scale. - Computation. This stage aimed at calculating the value of output
The final aim of this work is the development of a decision variables as a function of input variable values, under the
support system (DSS), based on techno-economic and environ- defined DOE constraints. Computation was carried out using
mental analysis, that identifies the system layout and plant size to ModeFRONTIER® software (Esteco, Trieste, Italy). The solutions
support fruit and vegetable producers in waste valorisation actions. calculated by the software represented the possible scenarios of
To this aim, the study-case of lettuce waste was taken into con- lettuce waste valorisation.
siderations, since this waste was successfully valorised on a labo- - Scheduling. The objective of this stage was to order the obtained
ratory scale by using both traditional and innovative technologies. scenarios according to the value of the output variables.
In particular, ready-to-drink juices, antioxidant extracts, functional Scheduling was carried out using MatLab® software (MATLAB
flour and a biodegradable expanded material were obtained by R2017a, 64-bit; The Mathworks Inc).
 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435 3

Fig. 1. Structure of the decision support system.

The scenario analysis represents the last phase, where the ob- In the paper, for the treatment of this waste, an industrial park
tained scenarios were compared and discussed in the light of integrating traditional lettuce waste management strategies with
multiple objectives. In particular, the study aimed at the max- innovative valorisation options was proposed. Based on informa-
imisation of environmental advantage indexes and at the mini- tion collected from the producers, lettuce waste is commonly
misation of economic effort indexes (as defined in paragraph 2.2). subjected to: (i) anaerobic digestion to produce digestate (fertil-
izer), biogas and, in turn, energy (by means of the cogeneration
2.2. Waste treatment plant model unit) (Garcia-Pen ~ a et al., 2011); (ii) composting to produce fertil-
izers (Himanen and Ha €nninen, 2011); (iii) carbonisation to produce
To produce value-added derivatives intended for food use, let- biocarbon (Li et al., 2019). According to recent studies on the
tuce waste is required to present a high homogeneity level. In innovative valorisation of lettuce waste, the latter could be sub-
addition, waste generation sites should not be very scattered, to jected to: (i) blanching and high pressure homogenisation to pro-
facilitate the collection and thus cut both collection and transport duce fresh juices (Plazzotta and Manzocco, 2019); (ii) ultrasound-
costs (Galanakis, 2012). For these reasons, this work was focused on assisted extraction to produce antioxidant polyphenolic extracts
lettuce waste generated in the food processing stage, that can (Plazzotta and Manzocco, 2018); (iii) air-drying and grinding to
ensure both a high compositional homogeneity and large amount produce vegetable flour intended for functional bakery products
in a reduced number of locations (i.e. the industrial plants). The first (Plazzotta et al., 2018a, 2018c); (iv) water substitution with ethanol
step was thus the collection of data relevant to the amount of let- and supercritical-CO2-drying to produce biodegradable expanded
tuce waste generated during fresh-cut processing. Official data materials for packaging or solvent adsorption applications
report that in Italy the fresh-cut lettuce market (ML) amounts up to (Plazzotta et al., 2018). In addition, side activities for the purifica-
about 105,300 tons/year (Confcoperative, 2016). Of that, 60% is tion and recycling of spent resources such as ethanol residue and
represented by whole-head lettuces, mainly Iceberg lettuce (Casati wastewater were hypothesized. The processes involved in tradi-
and Baldi, 2012). A survey conducted in a large Italian fresh-cut tional lettuce waste management, in its innovative valorisation and
company revealed that at least 35% of lettuce head weight is in the side activities of the industrial park are reported in Table 1,
wasted (%WL), mainly due to initial operations of external leaves and along with process yields. Real industrial processes were consid-
core removal (Plazzotta et al., 2017). Based on these data, the total ered for the process design of traditional waste management
amount of waste generated in 1 year in Italy by the fresh-cut pro- strategies (i.e. composting, anaerobic digestion and carbonisation)
cessing of whole-head lettuce (WL) was quantified in about 23,000 and side activities (i.e. wastewater treatment, ethanol recycling).
tons (eq. (1)). Such processes, in fact, present high technological readiness levels
(TRL), being actual processes, already applied on an industrial scale
WL ¼ 0:60  ML  %WL eq. 1 (European Commission Decision, 2017). By contrast, innovative
lettuce waste valorisation strategies, based on the production of
Similarly, the total amount of whole-head lettuce waste gener-
functional beverages, antioxidant extracts, vegetable flour and
ated by the large fresh cut company considered in the survey was
biodegradable materials by means of innovative technologies,
evaluated. In this company, about 20,000 tons of lettuce are pro-
present a low TRL. For this reason, process design was based on
cessed into fresh-cut derivatives. Considering 60% of the total
processes carried out on a laboratory scale and escalation factors of
processed lettuce to be represented by whole-head lettuces and
similar existing plants (Cristo  bal et al., 2018). The yields of each
35% waste production, the company would manage every year
lettuce waste process were estimated as % ratio of final output as
about 4,200 tons of whole-head lettuce waste.
4 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435

Table 1
Yields and outputs of processes involved in traditional management and innovative valorisation of lettuce waste and in related side activities. Output intended use and their
usage as marketable products or as resources recycled in the industrial park are also reported.

Strategy Process Yield (%) Output Intended use Usage

Traditional management Anaerobic digestion 3 Biogas Fuel for cogeneration Recycled

Cogeneration 60 Pure methane Energy Recycled
Composting 30 Fertilizer Fertilizer Recycled
Carbonisation 10 Biocarbon Fuel Market
Innovative valorisation Preliminary operations <50 Selected lettuce waste Raw material for valorisation strategies Recycled
Bioactive extraction 80 Lettuce bioactive extract Food supplement Market
Homogenisation 85 Lettuce homogenate Ready-to-eat soups and juice blends Market
Flour production 5 Lettuce flour Functional bakery products Market
Supercritical-CO2-drying 5 Lettuce material Biodegradable expanded material for packaging applications Market
Side activities Ethanol recycling 80 Ethanol Resource for industrial facilities Recycled
Carbon dioxide recycling 80 Carbon dioxide Resource for supercritical-CO2-drying Recycled
Wastewater treatment 60 Water Resource for industrial facilities Recycled

compared to the initial amount of raw materials entering the pro- The hypothesized industrial park is represented in Fig. 2, where
cess (Table 1). To this aim, industrial yields of traditional lettuce the flow diagram of the different processes involved in traditional
waste management options were retrieved from relevant literature lettuce waste management options, innovative valorisation strate-
(Keeling et al., 2003; Rossi and Bientinesi, 2016). By contrast, in the gies, and side activities, as well as their interactions are reported.
case of innovative valorisation strategies, laboratory results were In the case of traditional management options, lettuce waste
scaled up under the assumption that the same yields and perfor- would be straight directed to the proper industrial facility. By
mances would be obtained on an industrial scale, given the same contrast, the implementation of the innovative valorisation stra-
processing conditions (Albarelli et al., 2016). For example, the yield tegies would require a preliminary selection of lettuce waste, to
of air-drying and supercritical-CO2-drying resulted of 5%, due to remove spoiled and bruised parts. The latter would be managed by
95% moisture content of lettuce waste (Plazzotta et al., 2018). means of composting, anaerobic digestion or carbonisation. On the
Similarly, in the ultrasound assisted extraction of lettuce poly- contrary, the selected lettuce waste could be exploited as raw
phenols, about 20% of solid residue was retained in the filtration material to produce different valorisation outputs. It must be noted
step, leading to 80% yield (Plazzotta and Manzocco, 2018). that the need for lettuce waste selection introduces a high

Fig. 2. Flow diagram of resources (lettuce waste, ethanol, carbon dioxide, water and energy) in an industrial park integrating traditional management and innovative valorisation
strategies of lettuce waste.
 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435 5

uncertainty in the amount of lettuce waste available for innovative supply of plants and equipment present in the industrial park,
valorisation strategies, since the initial condition of lettuce waste respectively. Primary energy saving, (PES) was quantified based on
depends on unpredictable factors, such as weather and cultivation the biomethane-derived energy, obtained from lettuce waste
conditions. anaerobic digestion (eq. (4)):
Possible interactions among the different processing steps
involved in traditional and innovative valorisation strategies were PES ¼ CH4  xCH4 ½toe eq. 4
also identified. In fact, the integration of innovative strategies in the
existing waste management framework is surely most likely to where xCH4 is the standard coefficient for natural gas conversion
represent the real scenario of lettuce waste valorisation (Yu et al., into tons of oil equivalent (toe) (Simeoni et al., 2018) equal to
2019). In particular, the attention was focused on the possibility 8.2$104 (toe/Sm3CH4). The reduction of greenhouse gas emissions
to reduce the need for outsourcing of energy, water and raw ma- (CO2RED ) was also calculated from biogas-derived energy through
terial of a valorisation process by using the waste streams of other the following expression, using the proper emission conversion
processes integrated in the industrial park. factor of biomethane (Simeoni et al., 2018) (eq. (5)):

2.2.1. Energy demand

CO2RED ¼ CH4  mCH4 ½tons eq. 5
Data relevant to nominal energy demand of lettuce waste val-
where mCH4 is equal to 1.94$104 (tons 3
CO2/Sm CH4).
orisation plants, integrated into the designed industrial park, were
collected. Laboratory-scale data were directly derived from exper-
imental activity, while industrial-scale data were obtained from
company surveys. In particular, data relevant to traditional lettuce 2.2.3. Economic analysis
waste management strategies were collected from sector experts, Total investment cost (CI) and payback time (PB) were set as
engaged in the planning of local industrial activities. By contrast, in economic effort indexes of the designed industrial park.
the case of innovative valorisation strategies, that are not present in CI was calculated following eq. (6) as the sum of equipment cost
the current industrial practice, data were determined through (CE, V), cost attributed to civil work (CCW, V) and to plant design
scaling factors from similar existing plants and equipment (CPD, V).
(Cristo bal et al., 2018).
Collected data were elaborated to obtain energy functions, CI ¼ CE þ CPD þ CCW eq. 6
describing all the possibilities from a small laboratory scale up to
Since CE varies depending on the process scale, functions (in the
large industrial ones. Regression equations describing the variation
form of eqs. (2) and (3)), expressing CE as a function of equipment
of absorbed nominal power (Pi) as a function of maximum plant
power, were obtained and are reported in Supplementary Table S1.
capacity (tons of processed raw material or semi-finished product,
It must be noted that for low TRL technologies, cost estimation
WLi) were obtained and compared based on the correlation coef-
presents a ±30% accuracy, due to possible failures in inflation
ficient R2 (Microsoft® Excel 2016). Functions in the form of eqs. (2)
projection and cost growth, due to unpredictable events related to
and (3) were obtained and the equation presenting the highest R2
the high complex process and unproven technology (Tsagkari et al.,
was selected.
2016). According to sector experts’ opinion, CCW was set equal to 2/3
Pi ¼ mWLi þ q eq. 2 of CE and CPD was calculated as 2% of CE.
In addition to CE, the costs associated with daily operation and
maintenance of the industrial park (CO&M) were calculated ac-
Pi ¼ mlogðWLi Þ þ q eq. 3
cording to eq. (7).
These functions are reported in Supplementary Table S1. Such
functions allow estimating absorbed Pi of specific plants and CO&M ¼ CM þ CW þ CU þ CRM þ CWS eq. 7
equipment as a function of WLi, where i represents a generic
operation. Thus, they represent a flexible tool to describe a wide where CM (V) is the cost derived from CE.; CW (V) is the cost of
range of possible scenarios, according to the available lettuce waste workforce required for plant operation; CU (V) is the cost of utili-
amount. ties; CRM (V) is the cost of raw materials; CWS (V) is the cost of waste
stream management. A detailed explanation of each term of the eq.
2.2.2. Environmental advantage (7) is reported in the Supplementary Annex 1, along with sup-
Energy saving and consequent reduction of greenhouse gas porting references. Moreover, Supplementary Tables S2 and S3
emissions were set as indexes of the environmental advantage of show the cost values considered for workforce, raw materials and
the designed lettuce waste valorisation industrial park. This was utilities.
attributed to the biogas produced form anaerobic digestion of let- PB was calculated according to eq. (8), as the ratio of CI and the
tuce waste, which can be used as sustainable resource to partially annual net profit (CF):
fulfil the energy requirements of the industrial park, contributing to
PB ¼ CI =CF eq. 8
reduce the emissions of greenhouse gases. In addition, the recycle
of resources other than energy within the industrial park would CF was calculated based on eq. (8):
allow reducing the need for outsourcing. In this regard, Table 1
reports the main outputs of the lettuce waste processes, underly- CF ¼ R þ TEE  CO&M eq. 9
ing their potential recycle within the industrial park, instead of
being sold in the market. As an example, the carbon dioxide where R (V) are the revenues obtained from selling the valorisation
deriving from the co-generation unit involved in the conversion of products in the market, TEE are the “White Certificate” incentives
biogas from anaerobic digestion in methane, could be used in the (V) and CO&M is the operation and maintenance cost (V).
supercritical-CO2-drying of lettuce waste. Moreover, the digestate TEE incentives for saved energy were considered as possible
and the biogas-based energy deriving from anaerobic digestion sources of economic revenues (Oikonomou et al., 2009) and
could be entirely recycled for lettuce cultivation and electrical calculated following eq. (9).
6 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435

(Fig. 3,b). In these diagrams a box plot of a low level and a high level,
TEE ¼ PES x VTEE eq. 10 indicated respectively with “-” and “þ”, of each factor variable are
represented. A factor is important if it leads to a significant differ-
where VTEE (V) is the value of the incentives, based on current
ence between the means of the two groups, connected by a line.
updates (GME-GSE, 2018).
The higher the difference, the more important is the factor. The
In order to calculate the value of R, the outputs of both tradi-
connecting line can have either a positive or a negative slope: a
tional and innovative lettuce waste management options were
positive slope indicates a positive correlation between the factor
individuated, along with their yield, intended use (Table 1), and
and the response, a negative slope indicates a negative correlation.
unit price range. The output price range (p) was based on that of
As it can be noticed from the graphs, as the waste fraction delivered
corresponding products on the market, and is reported in Table 2.
to anaerobic digestion increases, the CO2 emission reduction in-
creases (mean value of 163 tons CO2/year, reaching a maximum
2.3. Simulation system design value of 208 tons CO2/year), thus leading to primary energy saving.
The waste fraction subjected to innovative valorisation has a similar
Aiming to optimize the treatment and valorisation plant but less pronounced effect, since its related connecting line has a
configuration, different designs have been investigated. An over- lower positive slope (mean value of 102 tons CO2/year, reaching a
view of the analysed DOE, describing variable range variation and maximum value of 140 tons CO2/year). This effect is due to the fact
incremental steps, is reported in Table 2. As explained in paragraph that lettuce waste valorisation leads to the production of innovative
2.2, lettuce waste availability for a typical Italian company has been sustainable derivatives, which have a reduced environmental
estimated up to 4,200 tons/year. Nevertheless, WL has been tested impact as compared to traditional ones. On the contrary, an in-
on a wider range, ranging from 3,600 to 8,400 tons/year. This crease in carbonisation and composting fractions leads to a
allowed taking into account the variations of waste amount, which decrease of the CO2 emission reduction of the plant, as it is shown
can vary according to unpredictable conditions, including weather, from the negative slope of the connecting lines.
cultivation yield, pests. Regarding the effects on the considered variables on the eco-
The fractions of waste material delivered to the different oper- nomic performances, as the waste fraction delivered to innovative
ations involved in traditional treatments and innovative valor- valorisation increases, the PB decreases, while an increase in the
isation technologies, ranging from 0 to 1, have been used to waste fraction subjected to traditional treatment fractions leads to
optimize the industrial park configuration. A variable value, ranging a higher PB. This is due to the higher added value and thus market
from 0 to 300 V, was set for incentives (VTEE), based on most recent price of the products obtained by innovative valorisation strategies,
updates (GME-GSE, 2018). As already anticipated, Table 1 reports as compared to those derived from traditional ones (Table 2).
the identified marketable outputs of lettuce waste valorisation
strategies, along with their intended use. Table 2 reports the price 3.1. MCDM scenarios analysis
(p) range of these valorisation products, which was defined based
on corresponding market products. In this regard, the choice to use Table 3 reports four of the scenarios among those obtained by
a price range rather than a medium price was based on the extreme the study. These scenarios were selected based on the achievement
variability of their values over time (Cristo  bal et al., 2018; Giraudet of each one of the objectives of the study, i.e. the maximisation of
et al., 2011). the environmental advantage and the minimisation of the eco-
Considering the high number of considered variables and nomic effort indexes of the lettuce waste valorisation industrial
deriving possible designs, a population of 500 individuals and 250 park. In these scenarios, the amount of whole-head lettuce waste
generation were chosen. The Design phase computed a total of produced during 1 year from a large fresh-cut company (about
125,000 possible scenarios. The latter were then scheduled ac- 4,200 tons, as discussed in paragraph 2.2) was considered. As ex-
cording to the values assumed by the environmental advantage and pected, the scenario allowing to maximise the environmental
economic effort indexes of the multi-objective study, as defined in advantage would be the one managing the major part (60%) of
paragraph 2.2. lettuce waste through anaerobic digestion to produce biogas. The
remaining lettuce waste fraction would be valorised into fresh
3. Results and discussion homogenates, antioxidant extracts and innovative biodegradable
materials. However, the CI of this scenario would be of 9.7 million V
The multi-objective optimization problem for the case study and with a PB higher than 2 years (Table 3). This can be attributed to
described in Chapter 2 has been solved as explained in Fig. 1 using a the high cost of equipment required for implementing innovative
16 GB RAM, i7 4770 3.40 GHz PC. technologies such as high pressure homogenisation, ultrasound
For the considered waste treatment and valorisation scenarios assisted extraction and supercritical-CO2-drying. Accordingly, the
the charts in Fig. 3 represent the main effect of decision variables on minimisation of CI would be reached by managing at least 90% of
the maximisation of CO2RED (Fig. 3,a) and the minimisation of PB lettuce waste through traditional options, not allowing a proper
valorisation of its rich composition. Moreover, this scenario would
also present limited environmental advantages and a PB longer
Table 2
than 3 years (Table 3). The latter would be minimised to just 4
Overview of the analysed DOE, reporting variable range variation and incremental
steps. months by valorising 70% lettuce waste into bioactive extracts. This
valorisation strategy, in fact, would highly increase the value chain
Variables Range of variation Incremental step
of lettuce waste, by producing a high-price food supplement
WL ton/year 3600 ÷ 8400 600 (Table 1, Table 2). Nevertheless, CI would be higher than 10.5 million
Operation fractions % 0÷1 0.1
V, while CO2RED and PES would still be half that those realised in the
VTEE V/TEE 0 ÷ 300 100
pbiochar V/kg 0.2 ÷ 0.9 0.05 scenario maximising the environmental advantage of the designed
pfood, supplements V/kg 4 ÷ 18 1 industrial park (Table 3). Therefore, all the scenarios reaching one of
phomogenized V/kg 1.15 ÷ 6.15 0.5 the study objectives would present some drawbacks. In this regard,
pflour V/kg 0.16 ÷ 1.6 0.16 the selection of a specific scenario should be driven by a compro-
pchips V/kg 0.03 ÷ 0.18 0.03
mise among the defined economic and environmental objectives.
 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435 7

Fig. 3. Main effects charts of carbonisation, composting, anaerobic digestion and valorisation on: a) CO2RED and b) PB. (For interpretation of the references to colour in this figure
legend, the reader is referred to the Web version of this article.)

Table 3
Possible scenarios of lettuce waste valorisation, according to the main study objectives.

Objective Processed waste (%, w/w) Reduction Primary Investment Payback

of energy cost (CI, V) time (PB,
Traditional Innovative
greenhouse saving years)
management valorisation
gas emissions (PES, toe/year)
Carbonisa-tion Compo-sting Anaerobic Lettuce Lettuce Lettuce Lettuce (CO2RED, tons
digestion flour homogenate bioactive material CO2/year)
extract

Maximisation of environmental 0 0 60 0 4 24 12 124.4 55.1 9,667,276 2.4

advantage
Minimisation of investment cost 70 0 20 0 1 9 0 39.1 17.3 8,502,699 3.1
Minimisation of pay-back time 20 10 0 0 0 70 0 63.1 28.0 10,535,299 0.3
Compromise 0 40 30 0 12 18 0 71.8 31.8 9,120,427 1.0

For this reason, a further scenario, deriving from a compromise waste would be valorised by the application of innovative valor-
solution, is presented in Table 3. In this scenario, 30% of lettuce isation strategies, based on the production of value-added
8 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435

homogenates and bioactive extracts using high pressure homoge- DSS Decision support system
nization and ultrasound-assisted extraction, presenting a CI lower FVW Fruit and vegetable waste
than 9.1 million V and a PB of about 1 year. The remaining 70% MCDM Multi criteria decision making
lettuce waste would be subjected to traditional management op- MFA Material flow analysis
tions (composting and anaerobic digestion), contributing to PB Payback time
greenhouse gases emission reduction and energy saving of about PES Primary energy saving
72 tons CO2/year and 32 toe/year respectively. TEE White certificates
TRL Technological readiness levels
4. Conclusions C Cost
P Power
In this study, a decision support system (DSS), based on techno- R Revenues
economic and environmental analysis was applied to support let- V Value
tuce producers in waste valorisation actions. The proposed method W Waste
was demonstrated to be highly flexible, since considering a variable x Primary energy conversion factor
range of waste amount, equipment cost, energy demand, and plant m Carbon dioxide emission conversion factor
productive capacity. It also allowed considering the integration of % Operation fraction
innovative valorisation pathways in the existing waste manage- Biochar Biochar related
ment system, towards a multiple “zero-waste” biorefinery concept. CH4 Methane related
The application of the proposed method supports the decision- Chips Chips related
maker into the identification of different rational solutions aiming CW Civil work
at maximisation of a specific environmental or economic objective, E Equipment
or to the identification of a compromise among the different sus- Flour Flour related
tainability objectives. In particular, the results obtained in this Food_supplement Food supplement related
study highlighted that the increase in lettuce waste fraction sub- Homogenized Homogenized related
jected to anaerobic digestion leads to the minimisation in primary i Generic operation
energy demand, and thus of CO2 emissions. By contrast, the I Investment
application of innovative valorisation strategies were shown to L Lettuce
maximise the economic performances of the considered system. M Maintenance
Nevertheless, the DSS model allows identifying compromise sce- O&M Operation and maintenance
narios, presenting good environmental and economic perfor- PD Plant design
mances. In the present study, transport cost and environmental RED Reduction
impact were considered negligible, due to industrial park-oriented RM Raw materials
view approach. Nevertheless, further studies on the sustainability U Utilities
assessment of lettuce waste management and valorisation could W Workforce
compute the economic and environmental impact of transport. WS Waste stream
The acquired results would allow the generation of a flexible
decision support tool to guide stakeholders’ and policy makers’ Appendix A. Supplementary data
investment in the most sustainable waste valorisation strategies.
This tool could be also exploited for promoting advantageous in- Supplementary data to this article can be found online at
dustrial symbiosis opportunities in the waste management sector. https://doi.org/10.1016/j.jclepro.2020.121435.

Funding
References
This research did not receive any specific grant from funding
2006/12/EC. Directive of the European parliament and council (EC) 12/2006 on
agencies in the public, commercial, or not-for-profit sectors. waste. Off. J. Eur. Commun., L114/9e21.
Albarelli, J.Q., Santos, D.T., Cocero, M.J., Meireles, M.A.A., 2016. Economic analysis of
Declaration of competing interest an integrated annatto seeds-sugarcane biorefinery using supercritical CO2
extraction as a first step. Materials 9, 1e19.
Alongi, M., Melchior, S., Anese, M., 2019. Reducing the glycemic index of short
The authors declare that they have no known competing dough biscuits by using apple pomace as a functional ingredient. LWT-Food Sci.
financial interests or personal relationships that could have Technol. 100, 300e305.
Casati, D., Baldi, L., 2012. Il confezionato tira nonostante la crisi. Terra e Vita 27,
appeared to influence the work reported in this paper. 34e37.
Confcoperative, 2016. Le insalate di quarta gamma, mercati e tendenze. Retrieved
CRediT authorship contribution statement October 3, 2018, from. http://www.internazionalizzazione.confcooperative.it/
LE-NEWS/ArtMID/500/ArticleID/468/Export-Mercati-Le-insalate-di-quarta-
gamma-mercati-e-tendenze.
Stella Plazzotta: Investigation, Formal analysis, Writing - orig- bal, J., Caldeira, C., Corrado, S., Sala, S., 2018. Techno-economic and profit-
Cristo
inal draft, Writing - review & editing. Mattia Cottes: Investigation, ability analysis of food waste biorefineries at European level. Bioresour. Technol.
259, 244e252.
Formal analysis, Writing - original draft, Writing - review & editing. European Commission Decision, 2017. Of 24 April 2017. Horizon 2020 Work Pro-
Patrizia Simeoni: Conceptualization, Writing - review & editing, gramme 2016e 2017 - 20. General Annexes, p. 2468. Retrieved from. https://ec.
Supervision. Lara Manzocco: Conceptualization, Writing - review europa.eu/research/participants/data/ref/h2020/other/wp/2016-2017/annexes/
h2020-wp1617-annex-ga_en.pdf. last accessed 11/01/2020.
& editing, Supervision.
FAO, 2014. Definitional framework of food losses and waste. FAO, Rome, Italy.
Ferreira, M.S.L., Santos, M.C.P., Moro, T.M.A., Basto, G.J., Andrade, R.M.S.,
Nomenclature 
Gonçalves, E.C.B.A., 2015. Formulation and characterization of functional foods
based on fruit and vegetable residue flour. J. Food Sci. Technol. 52, 822e830.
Galanakis, C.M., 2012. Recovery of high added-value components from food wastes:
CF Annual net profit conventional, emerging technologies and commercialized applications. Trends
DOE Design of experiment Food Sci. Technol. 26, 68e87.
 S. Plazzotta et al. / Journal of Cleaner Production 262 (2020) 121435 9

Galanakis, C.M., 2013. Emerging technologies for the production of nutraceuticals Papargyropoulou, E., Lozano, R., Steinberger, K.J., Wright, N., Ujang, Z. Bin, 2014. The
from agricultural by-products: a viewpoint of opportunities and challenges. food waste hierarchy as a framework for the management of food surplus and
Food Bioprod. Process. 91, 575e579. food waste. J. Clean. Prod. 76, 106e115.
Garcia-Pen ~ a, E.I., Parameswaran, P., Kang, D.W., Canul-Chan, M., Krajmalnik- Pfaltzgraff, L.A., De Bruyn, M., Cooper, E.C., Budarin, V., Clark, J.H., 2013. Food waste
Brown, R., 2011. Anaerobic digestion and co-digestion processes of vegetable biomass: a resource for high-value chemicals. Green Chem. 15, 307e314.
and fruit residues : process and microbial ecology. Biores. Technol. J. 102, Plazzotta, S., Calligaris, S., Manzocco, L., 2018b. Innovative bioaerogel-like materials
9447e9455. from fresh-cut salad waste via supercritical-CO2-drying. Innovat. Food Sci.
Giraudet, L.G., Bodineau, L., Finon, D., 2011. The costs and benefits of white certif- Emerg. Technol. 47, 485e492.
icates schemes. Energy Effic. 5, 179e199. Plazzotta, S., Manzocco, L., 2018a. Effect of ultrasounds and high pressure homog-
GME-GSE, 2018. Aggiornamento valore certificati bianchi o TEE. Retrieved October enization on the extraction of antioxidant polyphenols from lettuce waste.
15, 2018, from. https://www.ri-esco.it/aggiornamento-valore-certificati- Innovat. Food Sci. Emerg. Technol. 11e19.
bianchi-tee/#Sessioni_di_Mercato_TEE_Ottobre_2018. Plazzotta, S., Manzocco, L., 2019. High-pressure homogenisation combined with
Goula, A.M., Lazarides, H.N., 2015. Integrated processes can turn industrial food blanching to turn lettuce waste into a physically stable juice. Innovat. Food Sci.
waste into valuable food by-products and/or ingredients: the cases of olive mill Emerg. Technol. 52, 136e144.
and pomegranate wastes. J. Food Eng. 167, 45e50. Plazzotta, S., Manzocco, L., Nicoli, M.C., 2017. Fruit and vegetable waste manage-
Gustavsson, J., Cederberg, C., Sonesson, U., van Otterdijk, R., Meybeck, A., 2011. ment and the challenge of fresh-cut salad. Trends Food Sci. Technol. 63, 51e59.
Global food losses and waste: extent, causes and prevention. Food and Agri- Plazzotta, S., Calligaris, S., Manzocco, L., 2018a. Application of different drying
culture Organization of the United Nations, Rome, Italy. techniques to fresh-cut salad waste to obtain food ingredients rich in antioxi-
Heck, S., Rogers, M., 2014. Century resource revolution: how to capture the biggest dants and with high solvent loading capacity. LWT-Food Sci. Technol. 89,
business opportunity in a century. McKinsey & Company, Seattle. 276e283.
Himanen, M., Ha €nninen, K., 2011. Composting of bio-waste, aerobic and anaerobic Plazzotta, S., Sillani, S., Manzocco, L., 2018c. Exploitation of lettuce waste flour to
sludges - effect of feedstock on the process and quality of compost. Bioresour. increase bread functionality: effect on physical, nutritional, sensory properties
Technol. 102, 2842e2852. and on consumer response. Int. J. Food Sci. Technol. 1, 1e8.
Kader, A.A., 2005. Increasing food availability by reducing postharvest losses of Rossi, C., Bientinesi, I., 2016. Linee guida per realizzare impianti per la produzione di
fresh produce. Acta Hortic. 682, 2169e2175. biogas/biometano. Retrieved from. www.isaac-project.it.
Keeling, A.A., McCallum, K.R., Beckwith, C.P., 2003. Mature green waste compost Segre
, A., Falasconi, L., 2011. Il libro nero dello spreco in Italia: Il cibo. Ambiente
enhances growth and nitrogen uptake in wheat (Triticum aestivum L.) and Edizioni, Milan.
oilseed rape (Brassica napus L.) through the action of water-extractable factors. Sicaire, A.G., Abert, M., Fine, F., Carre , P., Tostain, S., Chemat, F., 2016. Ultrasound
Bioresour. Technol. 90, 127e132. induced green solvent extraction of oil from oleaginous seeds. Ultrason.
Lam, C.M., Yu, I.K.M., Hsu, S.C., Tsang, D.C.W., 2018. Life-cycle assessment on food Sonochem. 31, 319e329.
waste valorisation to value-added products. J. Clean. Prod. 199, 840e848. Simeoni, P., Nardin, G., Ciotti, G., 2018. Planning and design of sustainable smart
Li, Y., Liu, H., Xiao, K., Liu, X., Hu, H., Li, X., Yao, H., 2019. Correlations between the multi energy systems. The case of a food industrial district in Italy. Energy 163,
physicochemical properties of hydrochar and specific components of waste 443e456.
lettuce: influence of moisture, carbohydrates, proteins and lipids. Bioresour. Talens, C., Castro-Giraldez, M., Fito, P.J., 2016. A thermodynamic model for hot air
Technol. 272, 482e488. microwave drying of orange peel. J. Food Eng. 175, 33e42.
Mattiussi, A., Rosano, M., Simeoni, P., 2014. A decision support system for sustain- Tsagkari, M., Couturier, J., Kokossis, A., Dubois, J., 2016. Early-stage capital cost
able energy supply combining multi-objective and multi-attribute analysis: an estimation of biorefinery processes: a comparative study of heuristic tech-
Australian case study. Decis. Support Syst. 57, 150e159. niques. Chem. Sustain. 9, 2284e2297.
Meullemiestre, A., Breil, C., Abert-Vian, M., Chemat, F., 2016. Microwave, ultrasound, Xiong, X., Yu, I.K.M., Tsang, D.C.W., Bolan, N.S., Sik Ok, Y., Igalavithana, A.D., et al.,
thermal treatments, and bead milling as intensification techniques for extrac- 2019. Value-added chemicals from food supply chain wastes: state-of-the-art
tion of lipids from oleaginous Yarrowia lipolytica yeast for a biojetfuel appli- review and future prospects. Chem. Eng. J. 375, 121983.
cation. Bioresour. Technol. 211, 190e199. Yu, I.K.M., Attard, T.M., Chen, S.S., Tsang, D.C.W., Hunt, A.J., Jero^me, F., et al., 2019.
Oikonomou, V., Patel, M.K., Gaast, W. Van Der, Rietbergen, M., 2009. Voluntary Supercritical carbon dioxide extraction of value-added products and thermo-
agreements with white certificates for energy efficiency improvement as a chemical synthesis of platform chemicals from food waste. ACS Sustain. Chem.
hybrid policy instrument. Energy Pol. 37, 1970e1982. Eng. 7, 2821e2829.