When immediate interactive feedback boosts optimization problem solving: A 'human-in-the-loop' approach for solving Capacitated Vehicle Routing Problems.

In past, feedback in problem solving was found to improve human performance and focused mainly on learning applications. Interactive tools supporting decision-making and general problem-solving processes have long being developed to assist operations but not in optimization problem solving. Optimization problem solving is currently addressed within Operational Research (OR) through computational algorithms that aim to find the best solution in a problem (e.g. routing problem). Limited investigation there is on how computerized interactivity and metacognitive support (e.g. feedback and planning) can support optimization problem solving. This paper reports on human performance on Capacitated Vehicle Routing Problems (CVRPs) using paper-based problems and two different versions of an interactive computerized tool (one version with live explanatory and directive feedback alongside planning (strategy) support; one version without strategy support but with live explanatory feedback). Results suggest that human performance did not change when people were given paper-based post-problem feedback. On the contrary, participants' performance improved significantly when they used either version of the interactive tool that facilitated both live feedback support. No differences in performance across the two versions were observed. Implications on current theories and design implications for future optimization systems are discussed.