Assessing Values-based Sourcing Strategies in Regional Food Supply Networks : An Agent-based Approach
and unpredictable system-level behavior and outcomes ( e . g ., food hub success / failure ) that cannot be predicted by examining the motivations and behaviors of the individual participants ( Meter , 2006 ; Pathak , Day , Nair , Sawaya , & Kristal , 2007 ). Agent-based modeling ( ABM ) is a tool that is well-suited to capturing the complexity of such supply networks ( Choi , Dooley , & Rungtusanatham , 2001 ). For example , Krejci and Beamon ( 2015 ) developed a theoretical ABM to study the impact of farmer coordination on the development of regional food system structures and social sustainability outcomes . To gain an increased understanding of the preferences , drivers , attributes , and behaviors of food hub participants , as well as the factors that encourage / discourage consumers and producers to participate in the system over time , Krejci , Stone , Dorneich , and Gilbert ( 2016 ) developed an ABM of a regional food system in Iowa using NetLogo ( v . 5.0.2 ). The model was based on empirically derived inputs , which enabled a more realistic representation of the system and its constituent actors . This empirical ABM provides the basis for the study presented in this paper , in which the food hub manager ’ s sourcing strategy is investigated .
To collect the data for this study , a structured interview with consumer and producer members of an Iowa food hub was conducted to provide a scientific profile of both groups . These profiles would then be used to help identify critical variables and ultimately provide more accurate information to be used in modeling consumer and producer behavior . The interviews were conducted onsite at the food hub ’ s distribution center in Des Moines , Iowa . Interviewees first signed IRB-approved consent forms and then began a structured interview with a researcher for a period of 1 hour . Interviewers followed a strict interaction protocol so as to avoid influencing participants ’ responses . After completing all of the interview questions , participants were given the opportunity to ask questions of the researchers . Upon completion of the interviews , survey questions were transcribed into a spreadsheet and then categorized .
Data Analysis
In total , 33 individuals participated in the interview process ( 22 consumers and
11 producers ). The typical consumer was 48.5 years of age ( range 28 – 78 ), had a median household income of $ 100,000 ( range $ 40,000 –$ 300,000 ), and lived 6.5 miles ( SD = 3.2 ) from the food hub ( or associated distribution center ). They tended to be very comfortable with technology and were typically the primary shoppers for their family , which averaged 2.5 ( SD = 1 ) individuals . Consumers were also likely to have a college education . The typical producer was 49.1 years of age ( range 28 – 67 ), had a median household income of $ 78,000 ( range $ 13,000 – $ 175,000 ), and lived 39.8 miles ( SD = 22.7 ) from the food hub . Producers were very comfortable with technology , had a family which averaged 3.4 individuals
29