Multidisciplinary R&D project success in small firms: The role of multiproject status and project management experience

Abstract

R&D projects in small biotechnology firms frequently involve knowledge from multiple technical fields and research in different problem domains. An increase in project knowledge scope, defined as the number of technical fields an R&D project covers, can be challenging for resource-constrained small firms. These firms often rely primarily on their principal investigators (PIs), who act as heavyweight project managers in guiding project ideas to successful R&D outcomes. PIs also work concurrently on multiple projects, a strategy to promote learning across projects. To better understand how small firms PIs manage projects with high knowledge scope, our research assembles and analyzes a data set of 1374 R&D projects conducted by 933 small firms in the context of U.S. Small Business Administration awards. Results, after accounting for endogeneity, suggest a negative association between project knowledge scope and project success, which we measured using patent forward citation counts. We also find that a PI's multiproject status negatively moderates (i.e., amplifies) this association, while project management experience positively moderates (i.e., weakens) it. A follow-up post hoc analysis suggests that a shared problem domain is a key contingency for the moderation effects of both multiproject status and project management experience. Taken together, our research offers insights on how to effectively manage R&D projects in resourceconstrained small firms.

Vulnerability diffusions in software product networks

“…Similar results between different sets of hyperparameters are reported in other Node2Vec literature (Goyal & Ferrara, 2018;Grover & Leskovec, 2016). The adoption of time-dependent concepts for establishing the feature matrix of a node allows for better performance at predicting an edge using information about temporal edge relationships between time windows (Goyal & Ferrara, 2018;Nguyen & Tran, 2010). The remaining machine learning steps are the same as for the original Node2Vec (Grover & Leskovec, 2016).…”

“…Studies have developed theoretical foundations for understanding vulnerabilities in SPNs. Product defects and vulnerabilities emerge due to deficiencies in prior development iterations among the networked organizations (Browning & Yassine, 2016; Choo et al, 2020; Pan et al, 2022). Vulnerabilities in SPNs can occur continuously and transitively through common software development processes (Bensoussan et al, 2020).…”

The impact of crowdsourcing and user-driven innovation on R&D departments’ innovation activity: Application of multivariate correspondence analysis

“…Small entrepreneurs without an established history can find it difficult to access other forms of financing. Often, innovative companies obtain their first funding through crowdfunding platforms and can develop companies and produce products in a more professional manner (Brem et al, 2019;Pan et al, 2022). The idea of user-driven innovation, discussed in this article, is also connected with open-source communities.…”

Prediction of SMEs’ R&D performances by machine learning for project selection

“…Therefore, they have been limited in addressing the fundamental problems of expert evaluation discussed above. Meanwhile, there has been active empirical research on how to identify the various critical success factors (CSFs) that affect the success of R&D projects 22 – 25 . However, many of these success factors are difficult to quantify objectively, and most of the empirical studies have relied on the Likert scale to measure survey responses from experts 6 , 16 .…”