Modeling crowd-shippers’ decisions: a covariate-dependent choice model incorporating stated importance and certainty scores

With the rise of e-commerce and growing consumer expectations for rapid delivery, logistics providers are facing mounting pressure, leading to traffic congestion and greater environmental issues. In response, crowd-shipping (CS) has emerged as a promising solution to mitigate the environmental footprint of traditional logistics by leveraging existing trips to transport goods. The effectiveness of such service critically depends on citizens’ willingness to act as crowd-shippers. This study investigates the behavioral determinants of CS acceptance using an advanced stated choice experiment. A covariate-dependent choice model is estimated. Socio-demographics and stated importance (SI) scores of task attributes are used as covariates in the specification of random parameters to capture heterogeneity in the decision of CS acceptance. Results demonstrate that socio-demographics exhibit diverse influences depending on the nature of the attribute. For instance, higher-income individuals are more sensitive to economic-related factors, whereas privacy-related requirements are more strongly associated with gender differences. Individual perceived attribute importance plays a consistent role in shaping their decisions. Attributes (e.g., parcel size, goods fragility, incentives, and detours) that respondents consider important exert stronger average effects on acceptance choices, whereas those perceived as less important have diminished impacts. Moreover, the values of estimated taste parameters were reduced in magnitude after including certainty scores reported by respondents, suggesting that the barriers are less discouraging and incentives are less encouraging if certainty level is included in the estimation. Moreover, individuals with higher confidence for their choices require lower compensation for detours compared to the uncertain choices. This study demonstrates that incorporating socio-demographics, SI and certainty scores improve the reliability of preference estimation and offers richer behavioral insights. These findings help operators better identify key barriers and drivers of CS acceptance, provide practical guidance to develop flexible incentive schemes tailored to different travel contexts, and support the design of user-oriented CS systems.