Based on intrinsic bacterial characteristics and other external conditions, spoilage microbial organisms colonize and grow in specific food items at different stages in the food supply chain, which, in turn, leads to food loss and waste. Parameters that can influence the growth of spoilage microorganisms can be divided into (i) intrinsic and (ii) extrinsic parameters. (i) Intrinsic parameters are the physical and chemical properties of the food itself, and include water activity, pH, and availability of nutrients. (ii) Extrinsic parameters include environmental factors where food is stored, such as the temperature, availability of oxygen, and humidity (Veld, 1996; Sperber, 2009; Mena et al., 2014).

Sufficient nutrient composition, appropriate pH, and high water activity of meat and meat products make microbial spoilage very common in the meat industry.

Poultry spoilage is majorly driven by bacteria, though molds and yeasts also play active roles in the spoilage. Pseudomonas spp., including both fluorescent and non-fluorescent strains, are the prevalent spoilage bacteria in poultry (Morales et al., 2016). Surprisingly, Shewanella putrefaciens, which is commonly considered a marine bacterium, is a prevalent cause of poultry spoilage (Russell et al., 1995). Undesirable off-odors caused by amino acid metabolism, and the subsequent development of surface slime are the predominant features of poultry spoilage (Cerveny et al., 2009).

The spoilage of red meat is determined by its surface conditions (Wickramasinghe et al., 2019). The most common bacterial species responsible for red meat spoilage are Acinetobacter spp., Pseudomonas spp., and lactic acid bacteria (Mills et al., 2014). The key spoilage characteristics of red meat are off-odors, off-flavors, discoloration, and gas production (Sun and Holley, 2012). Dried meat surfaces favor the growth of psychrotrophic molds and yeast such as Rhizopus spp. and Torulopsis spp. (Mendonca, 2010). Major fungal spoilage-related defects in red meats are whiskers, black spot, white spot, the development of slime, and sticky surfaces (Alía et al., 2016).

Bacterial species implicated in processed meat and poultry spoilage are highly dependent on the presence of conducive conditions, such as nutrient (such as water) and oxygen availability. For example, spoilage in ground meat and poultry could be attributed to bacteria with varying oxygen requirements. For example, aerobic bacteria would thrive in ground meat and poultry due to the introduction of oxygen during the grinding and mixing process (Jay et al., 2003; Maćkiw et al., 2011; Rooney et al., 2011; Sharma et al., 2013).

Seafood is one of the most perishable foods due to its high water content, neutral pH, and adequate nutrients. Spoilage in seafood can be distinguished by a distinctive, unacceptable off-odor. Shewanella spp. and Pseudomonas spp., which could break down proteins in seafood and produce off-odors, are two bacterial species that prominently feature in the microbial spoilage of seafood (Wang et al., 2017, 2021). The specific species that could be responsible for spoilage varies according to extrinsic factors such as the temperature (Parlapani and Boziaris, 2016). A number of other bacteria, such as Carnobacterium spp. and Brochothrix spp. are also actively involved in the spoilage of seafood (Gennari et al., 1992; Laursen et al., 2006; Møretrø et al., 2016; Hoel et al., 2019).

Fresh and fresh-cut produce, which undergo minimal processing (with minimal alterations to their ‘fresh’ nature), could naturally carry microorganisms and thereby increase the probability of microbial spoilage (Harris et al., 2003).

The leading bacterial spoilage microorganism related to spoilage of fresh and fresh-cut vegetables is bacterial soft rot, caused by Erwinia carotovora and fluoresecent Pseudomonas spp. (Liao and Wells, 1987; Olsen et al., 2006; Shelake et al., 2022). Among the fungal sources of fresh and fresh-cut vegetable spoilage are gray rot caused by Botrytis cinerea (Romanazzi and Feliziani, 2014), Rhizopus soft rot caused by Rhizopus stolonifera (Singh and Afaque, 2021), and sour rot caused by Geotrichum candidum (Tournas, 2005).

The spoilage microbiota in fresh and fresh-cut fruits is largely dependent on the pH of the fruit. Melons, with a mildly acidic to neutral pH, are primarily impacted by E. carotovora and Pseudomonas spp., which also affect fresh and fresh-cut vegetables, depending on the storage temperature (Bruton et al., 1991; Ukuku et al., 2006). Fruits with low pH (<4.0) and relatively high concentrations of sugar like tree fruits, citrus fruits, and berries are particularly susceptible to spoilage by molds and yeast, while the growth of most bacteria is inhibited (Rawat, 2015). In fresh-cut fruits, on the other hand, yeast has been known to convert sugars in the fruit to CO₂ and ethanol, leading to spoilage (Barth et al., 2009). Molds such as R. stolonifer, B. cinerea, and Penicillium spp. such as P. expansum, P. digitatum, and P. italicum are capable of causing a broad range of soft-rot spoilage in fruits including Rhizopus soft rot, gray rot, blue mold rot, and green mold rot (Spotts et al., 1999; Elmer and Reglinski, 2006; Baggio et al., 2016).

Milk and dairy products are rich in water, fats, proteins, and vitamins that support the growth of diverse groups of microorganisms. Psychrotrophic bacteria, which can survive at low temperatures, are predominantly involved in the spoilage of milk and dairy products, which are commonly stored under refrigerated conditions.

Raw milk, which contains a near-complete nutritional profile and has a neutral pH, is an optimal place for the growth of a large variety of microorganisms. Gram negative bacteria make up a majority of the bacterial species contaminating raw milk. On the other hand, high amounts of gram-positive bacteria cause sourness in raw milk (Quigley et al., 2011). Molds and yeasts are capable of spoiling raw milk as well (Baur et al., 2015).

Pasteurized milk spoilage is typically caused by heat-stable bacteria such as Citrobacter spp. and the heat-stable lipolytic and proteolytic enzymes produced by Pseudomonas spp. (von Neubeck et al., 2015; Martin et al., 2018; Center for Dairy Research (CDR), 2020). These bacteria are characterized by the ability to survive under low temperatures, heat resistance capacity in proportion to their normal growth temperature, and the capacity to produce heat-stable extracellular enzymes such as proteases and lipases, which could result in undesirable off-odors and flavors (Reichler et al., 2019).

Cheese spoilage is typically caused by fungi, which can survive under low pH and low a_w. Cheese products are also susceptible to both gram-negative and gram-positive bacteria, which are responsible for early blowing (Alichanidis, 2007), off-odor, off-flavor, body defects (such as dryness, too much moisture, or mechanical holes), and pigment defects (Caldera et al., 2015) and gas defects in cheese (Le Bourhis et al., 2007; Ortakci et al., 2015).

Yogurt and other dairy products are susceptible to spoilage by a broad range of yeasts and molds. Yeasts are the most prevalent spoilage organism in yogurt and other dairy products. Yeasty odors, bitter flavors, and gas production are the key characteristics of fungal spoilage in yogurt and other dairy products (Ledenbach and Marshall, 2009).

Eggs are susceptible to spoilage by gram-negative bacteria and filamentous molds, even though they possess a system of barriers such as the shell, and the antimicrobials and high alkaline pH (7.6–9.2) of egg whites (Stadelman and Cotterill, 1995). Bacterial spoilage of eggs is characterized by “rots” (Jones and Musgrove, 2008; Fa Mansour et al., 2015). Another typical feature of bacterial spoilage of eggs is the characteristic sulfurous odor, associated with the metabolism of amino acids, which produce H₂S and odorous compounds (Jan et al., 2018). “Whiskers” and “pinspots” are the common features of mold spoilage on the surface and internal portions of eggs, respectively (Geornaras and Sofos, 2004).

Soil and water are primary sources of contamination in the pre-harvest, or farm, stages of the food supply chain. Soil, which houses both microorganisms such as bacteria, yeast, and mold, and vectors that carry these agents, is a direct contaminant for many food groups, including produce, cereals, and animal meat. Moreover, soil contains animal and bird feces, which act as reservoirs of microorganisms. Water acts as a vector for a number of bacteria, protozoans and parasites, and can contaminate foods during the harvest and pre-harvest stages (via irrigation; Ijabadeniyi et al., 2011). Additionally, wind carries mold, which causes mold spoilage of food. In animal and poultry operations, animal skin and animal feed, which also house many microorganisms, also act contaminating agents, especially when multiple animals are housed in close quarters (Sperber, 2009). Exposure of foods during these early stages, combined with environmental conditions that favor microbial growth (such as temperature), results in food spoilage or disease in animals/poultry. Eventually, spoiled food/diseased animals/birds would be discarded during the pre-harvest and harvest stages in order to protect and preserve others within the batch, pack, or flock.

The outer skin of fruits, vegetables, cereal grains, nuts and other plant-based foods, unless perforated in some way, is designed to keep out external contaminants, including those of microbial origin. Similarly, the skin and hide of meat animals are meant to protect the animal from microorganisms. Moreover, despite the presence of microorganisms in the intestinal tract of animals, the latter is essentially separated from the muscles and living tissues of the animal, leaving these parts sterile. However, processing tends to upset this fragile balance. Examples of processes that result in the sterility of the interior being compromised include removal of the bran from cereal grains, vigorous washing (particularly with contaminated water), chopping or juicing of fruits and vegetables, and removal of the hide and intestines (perforation of the intestine during this process could lead to very high contamination levels) and cutting and/or grinding of meat (which exposes a greater surface area of meat to microorganisms) (Sperber, 2009). These processes also expose food preparation, or food surface, areas to microorganisms, resulting in contamination and (cross)contamination of previously sterile food. Prior studies have reviewed the major causes of food loss in the processing module (Mena et al., 2014; Raak et al., 2017). Primarily, weather variations, poor post-harvest management, non-conformance with retailer or regulatory specifications, excessive waste during cutting or trimming, incorrect visual quality metrics (shape, size, etc.), and contamination or rot formation in the product were identified as causes of food being discarded. It is important to note that, although the exact proportion attributable to spoilage and contamination cannot yet be determined, deterioration in product quality due to (cross)contamination (of meat) and development of rots (in produce) remains a consistent factor contributing to food waste in the processing stages (Mena et al., 2014).

Human handling remains a chief cause of (cross)contamination in food processing facilities. The use of unclean hands or gloves, human physiological activities such as talking, coughing, or sneezing, movement between areas within the processing facility with different controls for microbial growth, as well as ineffective cleaning of food-handling utensils and equipment are some of the chief handling-related causes of microbial contamination, and eventually food waste (Sperber, 2009). An example of microbial (cross)contamination due to the use of unclean equipment is the outbreak of Salmonella from fresh-cut cantaloupes due to (cross)contamination with an unclean knife (Castillo et al., 2009). Water is another major point of contamination – the use of non-potable water for washing is a major cause of microbial spoilage, whereas improper storage of foods during cleaning and sanitation activities could result in direct contamination with aerosolized microorganisms from equipment nooks and crannies or the processing facility floors (Sperber, 2009). For example, the former was identified as the reason for a major Salmonella outbreak from fresh produce (oranges) in the U.S. (Parrish et al., 1997). This is also true for the spread and proliferation of microorganisms contributing to food spoilage across the food supply chain.

Food passing through the (required) transport and logistical stages in the supply chain is subject to mechanical damage, ranging from mild impacts and abrasions to severe structural damage. This, in turn, can promote microbial spoilage contamination and accelerate enzymatic degradation (Raak et al., 2017) in the damaged food item. Insufficient temperature management is another cause of microbial and physiological spoilage during transport. A high fraction of food loss and waste has been previously attributed to insufficient cold chain management – particularly during the summer months, when loss and wastage of fresh produce and meats tends to peak (Mena et al., 2014; Jedermann et al., 2014a; Hammond et al., 2015). However, the temperature at which foods, particularly fresh foods, are to be stored is also product-specific – for example, fresh tropical fruits would be susceptible to chilling injury and loss if transported under refrigeration temperatures (Jedermann et al., 2014c).

Food packaging acts as a primary barrier, protecting food from physical, chemical, and microbial contaminants. However, packaged foods are also susceptible to damage and spoilage from mishandling (causing physical defects), the use of already contaminated packaging material, or contact with a non-sterile environment (which can contain aerosolized microorganisms or spores) immediately prior to packaging. For example, a 1999 outbreak of Salmonella in the U.S. was attributed to unhygienic conditions where orange juice was mixed and bottled (Vojdani et al., 2008). Once packaged, food must be stored at the correct storage temperature – maintenance of a cold-chain across all stages of the food supply chain is extremely important, as described in section 3.4. It is also important to maintain and ensure cleanliness in all storage areas, particularly areas that are in direct contact with food, across the food supply chain. This is because unclean surfaces and areas can become significant contributors to cross-contamination, resulting in massive loss of food. For example, an inadvertent source of contamination during storage may be condensate formed in refrigeration units, which can carry microorganisms, and can be spread by the ventilation systems throughout the food-processing plants.

According to a survey of retail food service providers conducted by Teller et al. (2018) food waste at the retail level can be primarily attributed to lack of consumer demand, increased ordering of products, particularly seasonal products (related to forecasting of demand), and consumer intolerance to products that are slightly deformed or ‘ugly.’ The authors also attributed food waste to the range of brands available for a single product and the product(s) being too close to expiring at the time that they are delivered to the store. This was also validated by a descriptive analysis of the causes of food waste conducted by de Moraes et al. (2020). These authors also identified inappropriate storage, issues with transportation, and lack of stringent quality standards as drivers of spoilage-related food waste, although the percentage of waste attributed to these causes was relatively low (at a cumulative ~10%).

Although all sectors of the food supply chain and at the retail, food service, and consumer levels are involved in the generation of food waste, the high percentage of consumer-related waste necessitates the development and implementation of novel technologies to identify and potentially reduce food waste. Food packaging, in addition to providing essential product information to consumers, remains an essential method to ensure food safety, reduce enzyme and microbial activity, minimize exposure with the atmosphere, and overall increase shelf life (Poyatos-Racionero et al., 2018). Active and intelligent packaging technologies, particularly, could have a major impact on reducing food waste.

According to the European Union regulation EC/450/2009, intelligent packaging materials are those that monitor and indicate the condition of packaged food or the environment surrounding the food (Poyatos-Racionero et al., 2018). Traditionally, these have relied on the use of chemical indicators or coatings to communicate the characteristics of the food item to the processor, retailer, or consumer. For example, the chemical indicator could interact with food components or the metabolites in the headspace of packaged foods, or even with the storage environment, generating a visible response, such as a change in indicator color, which would correspond with the state of the food product. Recently, however, there has been an impetus to use more advanced indicators in smart packaging, such as radio frequency identification technologies (RFID), time–temperature indicators (TTIs), freshness indicators, chromogenic sensors, and global positioning system (GPS), which are increasingly being used to ensure food quality, fast communication, and contribute to better transport modalities and up-to-date information concerning shelf life (Raak et al., 2017; Poyatos-Racionero et al., 2018). Although currently available intelligent packaging systems have been extensively reviewed elsewhere (Vanderroost et al., 2014; Zhang et al., 2016), this study provides a brief overview of the most promising technologies to detect spoilage or quality issues in food, and thereby reduce food waste.

RFID tags are typically used to track and identify products, allowing for improved product traceability (Lee and Rahman, 2014), better inventory management and streamlined supply chain processes. RFID tags are particularly promising, since they allow storage of diverse streams of information, such as product source, environmental parameters, and expiration date. Such information would help ensure compliance with government regulations pertaining to food traceability (such as the Food Safety Modernization Act’s rule 204), Hazard Analysis and Risk Based Prevention Controls (HARPC), and cold chain management in a number of highly perishable food classes including fresh fruits and vegetables, frozen fish, and soft cheeses, while simultaneously allowing timely and accurate exchange of important information among trading partners. Examples of U.S. companies that have successfully implemented RFID technology for improved traceability, better cold chain management, demand management, and faster identification of products developed from contaminated or spoiled source material include Mission Foods, LaClare Farms, Chipotle, and Ste-Lor Oaks beef (Attaran, 2012; Littman, 2022). However, widespread implementation of this useful technology in the food industry has been impeded by high capital costs, lack of in-house expert knowledge, concerns regarding data privacy, and uncertainty regarding usage standards and regulations, among other issues (Attaran, 2012).

TTIs are highly efficient, relatively easy-to-operate indicators that continuously monitor the temperature of foods, particularly for refrigerated and frozen foods. TTIs are currently the most common system to identify physical, chemical, enzymatic, and microbial changes at a commercial level, primarily because they provide visual cues identifying changes in quality or freshness (Poyatos-Racionero et al., 2018), especially with major companies like 3M (3M™ MonitorMark™), the Cole-Parmer Instrument Company (Traceable ONE™, Traceable Excursion-Trac™, among others), and DeltaTRAK (WarmMark®) investing in developing novel, easy-to-decipher TTIs for the food industry (MarketWatch, 2023). However, these also have the potential disadvantages of spontaneous activation of chemicals, leading to false positive results; enzymatic instability; and leakage of the chemicals into the food product (Lin et al., 2021; Ye et al., 2022). Moreover, such technologies would not be cheap, since many of the available TTIs are not reusable.

Freshness indicators are another smart technology that could help detect freshness and quality of foods. These indicators highlight chemical changes occurring in food products during storage, such as changes in the concentration of metabolites like glucose, organic acids, carbon dioxide, biogenic amines, volatile nitrogen compounds or sulfur derivatives, which are potential signs of microbial growth (Arvanitoyannis and Stratakos, 2012; Poyatos-Racionero et al., 2018). Although a considerable amount of research has been conducted into freshness indicators such as ToxinGuard and SensorQ, and their efficacy in minimizing spoilage-related waste, there are currently very few that are commercially available (such as RipeSense®). It is important to note that a vast majority of these sensing technologies have been developed for use in foods packaged with modified atmosphere packaging (MAP) technology, which in itself was developed to increase shelf life and minimize microbial activity in packaged foods. However, their successful use in commercial applications remains rare. This could potentially be addressed by long-term validation of their safety (specifically for indicators that employ chemical components), improved regulatory mechanisms to allow their use in commercial applications, and increased research into the use of cost-saving materials in the development of these sensors.

Recent studies have also proposed combining intelligent packaging techniques with food date labels (in a manner similar to the schematic shown in Figure 2) as an effective method to provide consumers with real-time information concerning the quality and safety of food products, thereby reducing food waste at the consumer level. Adopting TTI into food date labels enables the estimation of remaining shelf-life in a non-destructive manner. Labuza and Szybist (2008) state that instead of traditional open-date labels, food products could be labeled as “use by XXX unless indicator shows …,” based on TTI design specifications. TTI labels not only offer insights about the freshness of the food product to consumers by indicating the “Use By” date, but also guarantee the safety of food and reduce food waste as they convey the condition of food products on a real-time basis by adopting TTI. A great amount of gas-sensitive smart labels such as oxygen-, CO₂-, and volatile compound-sensitive smart labels, have been reported to successfully convey up-to-the-minute information regarding the atmosphere in contact with food products (Saliu and Della Pergola, 2018; Wang et al., 2019; Niponsak et al., 2020). As such, gas-sensitive smart labels could be combined with conventional date labels (similar to TTI date labels) to help minimize food waste by supplying real-time data about the quality of food products. In addition, though not widespread, RFID smart labels have also demonstrated the potential to provide timely monitoring information for food products (Chen et al., 2023).

Although quantitative microbial risk assessment (QMRA) has been extensively applied to characterize the risk of foodborne illness, its application in food spoilage remains novel. However, over the past decade, studies have postulated on the importance of quantitative microbial spoilage risk assessment to develop more effective management strategies to minimize food spoilage, extend shelf life, and limit global food waste [European Food Safety Authority (EFSA), 2020]. Although the scope of any risk assessment would be to assess and manage the risk of the presence of a microbial agent in food, a QMSRA aims to assess both the growth of the spoilage microorganism of interest and its spoilage-associated metabolic activity. In essence, in addition to the growth response of the spoilage microorganism to prevalent environmental conditions such as the temperature, humidity, and oxygen content, this framework would have to incorporate models to predict the conditions that would lead to the development of off-flavors, odors, and textures, as well as the probability that a consumer would reject the food item based on their personal perception of spoilage (Koutsoumanis et al., 2021). Despite these differences, a few studies have developed comprehensive QMSRAs for various food-spoilage microorganism combinations. A majority of these risk assessments focus on quantifying spoilage risk from molds such as A. fischeri in pasteurized strawberry puree (Dos Santos et al., 2020), A. niger in yogurt (Gougouli and Koutsoumanis, 2017), and other molds in bread (Dagnas et al., 2017). Spoilage risk assessments have also been developed for bacterial spoilage agents such as Geobacillus stearothermophilus (canned green beans and bread; Rigaux et al., 2014; Pujol et al., 2015), as well as pathogenic agents responsible for food waste, such as Clostridium botulinum (ultra-high temperature pasteurized milk) and Bacillus cereus (bread; Pujol et al., 2015). However, these are few and far between, with a standardized framework for QMSRA being made available only as recently as 2021 (Koutsoumanis et al., 2021). The use of QMSRA would be particularly useful at the supply chain level, where the industry (producers and processors) can apply pre-set quality targets by developing and implementing effective control programs to minimize spoilage-related activities and thereby reduce food waste (Koutsoumanis et al., 2016). For example, QMSRA models could be used to simulate what-if scenarios with different combinations of product formulation, process and storage conditions, and packaging choices, which in turn could be used to assess the impact on spoilage microorganism growth and activity. If such models were to be adopted by food processors and retailers, they would help supplement the empirical decision making process currently being employed by managers in the food industry, particularly in selecting expiration and ‘best by’ dates that could minimize food waste at the retail and consumer levels (Koutsoumanis et al., 2021).

Other novel technologies that have recently been making waves in identifying or minimizing food spoilage include imaging technologies (such as hyperspectral imaging) to determine food freshness and quality in a non-invasive manner and biodegradable and edible food coatings such as Apeel that slow down the rate of water evaporation that causes fruits and vegetables to degrade while allowing for a natural exchange of gases with the atmosphere, which could be particularly useful in ensuring that produce reaches retail and the consumers before it spoils (Cosgrove, 2018; Milland and Taylor, 2018; Central Florida Store Services, 2020). Delaying produce ageing is another method being explored to extend produce shelf life and consequently minimize pre-retail food loss. The USDA-funded Hazel 100™ utilizes 1-methylcyclopropene (1-MCP) to reduce the production and absorption of ethylene, which is naturally produced by produce as it ages. This technology has proven to be effective in delaying ageing and reducing food loss in produce such as apples, cherries, limes, avocados, and melons (Fresh Fruit Portal, 2019; Diep, 2022). Hazel Technologies is also in the process of developing an antimicrobial-based technology (Hazel Endure™) to reduce mold spoilage in fruits such as grapes and berries (Hazel Technologies, 2023). In the meat industry, a novel nanofiber absorbent pad containing the antimicrobial polyhexamethylenebiguanide hydrochloride (PHMB) has been recently developed to reduce meat spoilage. This method combines antimicrobial technology with an anthocyanin-based indicator technology to easily detect meat spoilage. If this method proves to translate well into a commercial setting, it could have a significant impact on minimizing post-retail food waste (Jiao et al., 2023).

Another method that has seen growing interest by stakeholders in the food supply chain to reduce spoilage and food waste is IoT. Internet of Things or IoT is a platform allowing seamless communication between smart devices, enabling information sharing across platforms in a convenient manner (Vermesan et al., 2011; Marjani et al., 2017). In the context of the food supply chain, when a food product is assigned a digital footprint (such as one, or a series of, sensor(s), GPS technology, or cameras), the latter can be used to collect data on the food (humidity, oxygen content, quality metrics) and the surrounding environment (such as temperature and relative humidity) in the entire food supply chain (agricultural production, post-harvest handling, processing, transport, and storage) and beyond, as well as data on transport logistics and packaging. In this framework, information on the product can be collected using a range of sensors, such as temperature, humidity, chemical, and optical sensors. This data can then be processed for food quality monitoring using machine learning models such as clustering algorithms, decision tress, regression models, artificial neural networks, and support vector machine models (Ahmadzadeh et al., 2023). The results of these analyses can then be consolidated and utilized by the stakeholders to make critical decisions regarding strategies to handle food in order to minimize food loss and waste (Vermesan et al., 2011). However, there are a number of technical (lack of technical expertise), financial (high capital costs involved in setting up a system of sensors, for example), social (issues regarding data privacy and sharing), operational (lack of standards and communication protocols), educational (educating food service providers to switch from a well-established system to a newer system; educating the public), and governmental (lack of standardized regulations) challenges that work against the use of IoT in the food industry (Aamer et al., 2021; Ahmadzadeh et al., 2023).

Additional artificial intelligence (AI)-based applications to minimize food waste include the development of AI-based dynamic pricing to reduce the selling price of food closer to the “Sell by” or “Use by” date, or rotating food out based on a “first expired, first out” system, rather than the prevalent “first in, first out” system.