The study was conducted in the Peruvian provinces of Coronel Portillo and Padre Abad, located in the department of Ucayali (Figure 1). Covering an area of approximately 102 410 km2 (39 536 sq mi), the region is located in the eastern part of the country and encompasses portions of the Amazon basin. The region’s climate is typically hot and humid. The average annual precipitation is approximately 1500 mm in Coronel Portillo and 2000 mm in Padre Abad, with the dry season typically spanning from June to August.

Figure 1
Study area location in Ucayali region, Peru.

Two distinct ecosystems define the study area: the restinga, a floodable alluvial forest, and the high- altitude area, incorporating both low and high hill forest. The flood seasonality allows for the cultivation of crops with short vegetative periods such as rice, corn, peanuts, beans, cassava, banana, vegetables, and cucurbits. Additionally, crops such as camu camu (Myrciaria dubia), sapote (Pouteria sapota), and soursop (Annona muricata), which can withstand flooding and root saturation during high water levels, are well- suited for these zones. Conversely, the high-altitude areas are located on hills and terraces that are not subject to flooding.

Agriculture forms a primary economic pillar in the Ucayali region, conducted predominantly by family units around the department’s major rivers and main roads. As per data provided by the Statistics Office of the Regional Office of Agriculture, the region’s largest crop production in 2020 was oil palm (Elaeis guineensis), banana, cassava (Manihot esculenta), and rice (Oryza sativa) (Panduro Pisco, 2021).

study area houses a total of 112 agro-industrial enterprises engaged in processing a variety of inputs, according to data from the Regional Agrarian Directorate of the Department of Ucayali (DRA Ucayali, 2020) (Table S1). 57 agro-industrial enterprises were preliminary selected, employing Equation 1 with a sampling error of 10 %. Initial contact was made with companies that were geographically close to one another, easily accessible, and not located in remote areas far from roads or other companies. Of the 57 companies contacted, 21 were non-operational and 16 did not respond. A total of 20 companies were ultimately surveyed, but only 17 provided comprehensive information about raw material volumes, production, and solid waste generation.

With these final numbers, the margin of error was recalculated using Equation 1.

Where n represents the sample size, p is the probability of success (set at 87.7 % due to the rural setting and greater dispersion), q is the probability of failure (12.3 %), E is the margin of error, N is the population size (112), and Z is the confidence level (set at 1.96 for a confidence level of 1-a = 95 %). According to the final sample size, the margin of error was redetermined as 14 %.

Surveys were administered in October and November 2020 to companies involved in the processing of sugarcane (Saccharum officinarum), rice, cassava, plantain, coffee (Coffea arabica), cocoa, and palm oil. During each company visit, a structured survey (Table S2) comprising both closed and open-ended questions was used to gather data on their production, raw material consumption, waste generation, access to basic services, and environmental management practices. Operational managers provided flow charts data (Figure S1).

The rates of waste generation per feedstock and product, and feedstock efficiency, were calculated using Equations 2, 3 and 4, respectively.

Here, SW refers to the monthly volume of solid waste generated (in tons), F is the monthly volume of feedstock, and P is the monthly production volume (both in tons). Equations 5 and 6 were used to estimate waste and cellulose generation.

The term “production” in these equations refers to the yield of each crop in 2020, as per data from the Integrated System of Agricultural Statistics of the Peruvian Ministry of Agricultural Development and Irrigation (Midagri, 2021). The cellulose percentages were derived from references that specify the chemical composition of each waste type (Table 4).

The survey focuses on the agro-industrial activities in a specific area, highlighting the diversity in the scale of operations among enterprises. Larger enterprises, such as those producing alcohol, rice, and palm oil, possess extensive machinery and workforce. In contrast, smaller enterprises, often with less than ten workers, are engaged in producing various products like panela, aguardiente, and banana chips. Environmental management is notably lacking in small and medium-sized enterprises, with most perceiving environmental issues as a global rather than local concern. Only one company, Indomalsa, actively employs an environmental management tool. Energy consumption in these companies is primarily derived from non-electric sources, with diesel and wood being common, reflecting a pattern seen in broader agricultural sectors. However, there are disparities in electricity consumption costs among these enterprises, with those in rice production incurring significantly higher costs compared to those in cocoa processing and chocolate production, as detailed in the surveyed data.

Table 1 presents the production characteristics of the surveyed companies, providing detailed insight into the main product, total number of workers, total number of equipment units, frequency of equipment maintenance, and monthly electricity expenditure.

Table 1
Overview of production characteristics and operational details of agricultural and agro-industrial companies in Ucayali region, Peru

*Note: S.A.C. stands for Sociedad Anónima Cerrada, closed stock company, and E.I.R.L stands for Empresa Individual de Responsabilidad Limitada, individual limited liability corporation. They are part of the company’s legal names and typically not translated. The source of the equipment is denoted as ‘n’ for national and ‘i’ for international. The electricity supply is represented as ‘R’ for a combination of motor and solar panel power, while ‘E’ stands for power supplied by the regional electricity distribution company, Electro Ucayali.

The flow diagrams in supplementary Figure S1 depict the distinct processes and by-products of waste generated by the various companies surveyed in Table 1. These flow diagrams offer an exhaustive illustration of each process under investigation, detailing both the input and output components of each operation.

Table 2 collates essential information on each company’s primary feedstock, end products, and waste materials. In addition to this, the table also sheds light on the final destination of the waste produced.

Table 2
Comparative analysis of feedstock sources, products, main solid wastes, and waste management practices of agro-industrial companies in Ucayali region, Peru

*Note: The feedstock supply is indicated as follows: ‘o’ signifies sourcing from the company’s or association’s own land, ‘p’ denotes sourcing from partners-individuals who own land that is part of the company or association-, and ‘f’ represents sourcing from farmers, who are third parties not affiliated with the company or association. The distribution channels for the product are represented as ‘l’ for local, ‘n’ for national, and ‘i’ for international.

The study calculated solid waste generation rates from various raw materials and products to estimate the volume available for cellulose recovery (Table 3). Sugarcane, cassava, and oil palm were found to be less efficient, generating more waste due to their processing methods. Conversely, rice, banana, coffee, and cocoa were more efficient, producing higher yields with most waste comprising husks and shells.

Table 3
Assessment of feedstock quality, solid waste generation rates, and feedstock conversion efficiency for different agro-industrial products in Ucayali region, Peru

*Note: According to the Kruskal-Wallis test, the results are not statistically significantly different (waste rate per feedstock p= 0.087, waste rate per product p= 0.090, and feedstock efficiency p= 0.143).

Table 4 presents the estimated solid waste generation and cellulose pulp yield for each crop studied, based on national production figures. Sugarcane processed for sugar production showed the highest levels of waste and cellulose pulp generation among all the crops analyzed.

Table 4
Estimated solid waste generation and cellulose extraction potential from agricultural residues by Peruvian departments in 2020

*Note: The data for each crop’s production corresponds to the year 2020 and is sourced from the Integrated System of Agricultural Statistics of the Peruvian Ministry of Agrarian Development and Risk (Midagri, 2021). The potential for cellulose production was calculated with the following figures: between 40 % and 45 % in sugarcane (Singh et al., 2022), 35 % in coffee husk (Janissen and Huynh, 2018), 40.5 % in cassava husk (Widiarto et al., 2019), between 12 % and 59 % in banana husk (Tibolla et al., 2019), 35 % in cocoa husk (Lubis et al., 2018), between 33 % and 40 % in rice husk (Gao et al., 2018), and 33 % oil palm stalk (Neyra-Vasquez et al., 2022).

In smaller agro-industrial enterprises, operational sustainability is heavily reliant on the cohesion and partnership among members. The failure of many such enterprises is often due to internal conflicts and lack of strong organizational structures. Regarding health and safety, only a few companies have implemented comprehensive systems, with only Santa Clara and Moliexpress fully enforcing these programs. Notably, none of the companies surveyed have ISO 14001 certification for environmental management, although two of them, Landbar and Indomalsa, have ISO 9001 certification, indicating a commitment to quality management. The energy consumption in these enterprises is minimal, often due to the artisanal nature of their production processes which mainly utilize mechanical energy and gas, with some even reporting no energy costs. A few have low electricity costs due to reliance on artisanal methods, and one company has invested in solar panels to offset energy expenses, reflecting the small scale and low-tech operations prevalent in these firms.

Table 2 shows that sugarcane bagasse and rice husks are utilized for energy recovery through incineration. Furthermore, both moist and dry bagasse derived from sugarcane are repurposed as compost to enrich the soil for sugarcane cultivation. Similarly, cassava peels are utilized as compost for crop nutrition. Rice husks have a wide range of reuse applications, including as an ingredient mixed for small-scale fishing, as feed for pigs, and as bedding in chicken sheds, where they eventually decompose to form organic fertilizer. However, when these husks are not sold or incinerated, they are improperly disposed of in vacant plots. This poses an inconvenience for local residents when strong winds lift and scatter the husk dust.

Banana peels are repurposed as pig feed, and bananas deemed unfit for sale are consumed by workers involved in the selection process, often prepared as boiled green bananas.

Residual coffee husks are also repurposed as fertilizer. Nonetheless, this use has potential environmental implications due to the presence of caffeine, chlorogenic acid, and tannins, substances that can exert significant ecotoxicological effects on aquatic life forms (Marew et al., 2024). Cocoa husks are utilized both as fertilizers and as ingredients in infusions. Meanwhile, residues from palm oil processing can be put to various uses (Mora-Villalobos et al., 2023), such as a filling substance in soil or as material for graveling. The company surveyed utilizes the empty fruit bunches and mesocarp fiber for energy through burning. Furthermore, palm kernel waste, which is also abundant in the study area, can be utilized as a soil filler or for graveling.

The surveyed companies are actively engaged in waste reuse and repurposing initiatives, including composting, selling, and using waste for energy. However, there is room for improvement in managing non-reusable waste and mitigating health and environmental risks associated with waste reuse, highlighting the necessity for enhanced waste management practices.

Table 3 presents the average rates of solid waste generation and feedstock efficiency. As an illustration, for every ton of rice produced, 0,57 tons of rice husks are generated. In comparison, Spada et al. (2020) documented a waste generation rate of 0.23 tons per ton of rice produced. This discrepancy could be accounted for by differences in processing efficiency or variations in feedstock quality. For palm oil processing, a yield of 16 % was computed, which aligns with the findings by Dungani et al. (2018). This implies that the remainder of the processed materials is waste. Similarly, the yield of cassava processing for starch was recorded at 20 %, which is consistent with the findings of Ekop et al. (2019), who reported that every ton of fresh cassava processed results in 680 kg of total waste. However, this study only accounted for waste derived from cassava peels. Cocoa husk makes up 52 % to 75 % of the total wet weight of the fruit (Lubis et al., 2018). However, in this study, cocoa husk represents 16 % of the dry seed weight, suggesting an 84 % yield of waste. The quantity of sugarcane bagasse generated per ton of processed sugarcane aligns with the findings by Toscano Miranda et al. (2021), who calculated that each ton of sugarcane results in approximately 270 kg of sugarcane bagasse. However, other studies (Xu et al., 2018) have posited that sugarcane bagasse constitutes up to 50 % of the total volume of processed sugarcane. According to existing literature, the coffee industry generates between 500 kg and 600 kg of husks for every ton of processed coffee (Morales-Martínez et al., 2021), which is in line with this study. However, some research (Dadi et al., 2019) suggests a lower waste generation rate, stating that each ton of coffee produced results in just 1 kg of husk during the drying process.

The study highlights the potential for sustainable waste management in the agro-industrial sector, suggesting that agro-industrial waste could be repurposed in industries like paper, packaging, textiles, and biofuels. It emphasizes the need for national and local governments to adopt effective waste management and reuse strategies to foster a circular economy. The research underscores the importance of enforcing responsible waste disposal practices and recommends the creation of stringent environmental policies and regulations. These regulations would aim to reduce waste, increase energy efficiency, and promote waste reuse, transforming waste into a valuable resource. Additionally, the study suggests that building the capacity of small and medium- sized enterprises through training in partnership and environmental management is essential for implementing sustainable practices and achieving a more sustainable future in the agro-industrial sector.

While the primary focus of this study is on the extraction of cellulose for biodegradable packaging, recognizing the broader potential of agro-industrial waste can contribute significantly to sustainability efforts in the region. In addition to cellulose extraction, the study recommends exploring alternative waste valorization strategies for agro-industrial waste to maximize its potential. For instance, sugarcane bagasse and rice husks can be processed for biochar for soil enrichment or for bioenergy, such as biofuels, or biogas. Cassava peels, rich in starch, can be processed for bioplastics or for biogas production. Coffee and cocoa husks, which contain bioactive compounds, can be utilized in the pharmaceutical and cosmetic sectors, adding value to these by-products. Bananas peels, in addition, can be processed for use in animals’ supplements, or in pectin extraction for use in foods. The implementation of such options would enhance resource efficiency, create new economic opportunities, and significantly reduce environmental impacts, towards a more sustainable and circular agro-industrial economy in the region.