Foodborne disease outbreaks associated with produces have been increasing in occurrence worldwide. This study investigated microbial contamination levels on thirteen kinds of agricultural products from farms stage to evaluate potential hazards associated with foodborne illness. A total of 1,820 samples were collected in major cultivating area from 2013 through 2015, and analyzed to enumerate aerobic bacterial counts, coliforms/E. coli, Bacillus cereus and Staphylococcus aureus. In addition, the prevalence study for four kinds of microorganisms (Escherichia coli, E. coli O157:H7, Salmonella spp. and Listeria monocytogenes) was performed on each sample. Aerobic bacterial counts ranged from 0.01 to 7.18 log CFU/g, with the highest bacterial cell counts recorded for watermelon. Coliforms were detected in 651 samples (35.8%) with a minimum of 0.01 log CFU/g and a maximum of more than 5 log CFU/g. B. cereus was detected in 169 samples (9.3%) ranging from < 0.01 to 2.48 log CFU/g among total samples analyzed. S. aureus was detected in 14 samples (0.7%) with a minimum of 0.01 log CFU/g and a maximum of 1.69 log CFU/g. E. coli was detected in 101 samples (5.5%) among 1,820 samples. E. coli O157:H7, Salmonella spp. and L. monocytogenes were not detected in any of the samples. The microbial contamination levels of several agricultural products determined in this study may be used as the fundamental data for microbiological risk assessment (MRA).

This study was conducted to develop an agent-based computing platform enabling simulation of on-farm produce contamination by enteric foodborne pathogens, which is herein called PPMCS (Preharvest Produce Microbial Contamination Simulator). Also, fecal contamination of preharvest produce was simulated using PPMCS. Although Agent-based Modeling and Simulation, the tool applied in this study, is rather popular in where socio-economical human behaviors or ecological fate of animals in their niche are to be predicted, the incidence of on-farm produce contamination which are thought to be sporadic has never been simulated using this tool. The agents in PPMCS including crop, animal as a source of fecal contamination, and fly as a vector spreading the fecal contamination are given their intrinsic behaviors that are set to be executed at certain probability. Once all these agents are on-set following the intrinsic behavioral rules, consequences as the sum of all the behaviors in the system can be monitored real-time. When fecal contamination of preharvest produce was simulated in PPMCS as numbers of animals, flies, and initially contaminated plants change, the number of animals intruding cropping area affected most on the number of contaminated plants at harvest. For further application, the behaviors and variables of the agents are adjustable depending on user’s own scenario of interest. This feature allows PPMCS to be utilized in where different simulating conditions are tested.