Massive regression and forecasting tasks are generally cost-sensitive regression learning problems with asymmetric costs between over-prediction and under-prediction. However, existing classic methods, such as clustering and feature selection, are subject to difficulties in dealing with small datasets. As one of the key challenges, it is difficult to statistically validate the importance of features using traditional algorithms (e.g. the Boruta algorithm) owing to insufficient available data. By leveraging the feature information intra-cluster (item group with similar attributes), we propose an intra-cluster product favoured (ICPF) feature selection algorithm to select the information based on the traditional filtering method (specifically the Boruta algorithm in our study). The experimental results show that the ICPF algorithm significantly reduces the number of dimensions of the selected feature set and improves the performance of cost-sensitive regression learning. The misprediction cost decreased by 33.5% (linear-linear cost function) and 32.4% (quadratic-quadratic cost function) after adopting the ICPF algorithm. In addition, the advantage of the ICPF algorithm is robust to other regression models, such as random forest and XGboost.
Efficient business process is important to the operations of supply chain finance (SCF). Many deficiencies exist in the processes of SCF such as complicated workflows and high time-consuming steps. However, few studies have paid attention to evaluate and improve the performance of SCF processes. We empirically model and investigate the processes of supply chain finance by constructing a Stochastic Petri Net based on the field survey of a focal firm. Two critical indices, place busy rate and transition utilization rate, are evaluated. The results demonstrate that some places (transitions) of the Petri Net have high busy rates (utilization rates). By integrating the Petri Net and dependency graph, several key places and transitions in the Petri Net of SCF processes are identified for further optimization. To improve the performance of SCF processes, we propose three optional adjusting schemes on the basis of information sharing perspective (for the first half of the Petri Net) and risk sharing perspective (for the second half of the Petri Net). The proposed optimization strategies have further been proven to reduce the place busy rates, shorten the process, and improve the process efficiency of the supply chain finance.
Transactive memory system (TMS), as the powerful collective cognition, is well-documented to support team performance. Given today’s boundaryless era with great uncertainty, managing boundaries has become challenging but critical for teams. However, how TMS would influence teams to engage with external environments through boundary management is still underexplored. Adopting an open-system perspective, we explore how TMS influences teams’ boundary management and further contributes to team performance under uncertain environments. Empirical results from 337 respondents embedded in 56 teams revealed that TMS was positively related to all three types of team boundary management. Among them, boundary spanning, and reinforcement was positively related to team performance, while boundary reinforcement mediated the relationship between TMS and team performance. Lastly, environmental uncertainty was found to play an important moderating role in the TMS - team performance relationship. Accordingly, this research advances the knowledge of how teams would manage while being influenced by their external environments.