Fault-Tolerant, but Paradoxical Path-Finding in Physical and Conceptual Systems

TL;DR

This paper explores models for navigating unreliable networks using risk accumulation, revealing paradoxical behaviors in path selection that have implications for design and development processes.

Contribution

It introduces two novel risk-based path-finding models and uncovers paradoxical effects in their application to conceptual networks.

Findings

Most reliable path shifts from wider to narrower as risk increases.

Paradoxical path changes suggest novices should follow simpler routes.

Implications for designing processes in uncertain environments.

Abstract

We report our initial investigations into reliability and path-finding based models and propose future areas of interest. Inspired by broken sidewalks during on-campus construction projects, we develop two models for navigating this "unreliable network." These are based on a concept of "accumulating risk" backward from the destination, and both operate on directed acyclic graphs with a probability of failure associated with each edge. The first serves to introduce and has faults addressed by the second, more conservative model. Next, we show a paradox when these models are used to construct polynomials on conceptual networks, such as design processes and software development life cycles. When the risk of a network increases uniformly, the most reliable path changes from wider and longer to shorter and narrower. If we let professional inexperience--such as with entry level cooks and software developers--represent probability of edge failure, does this change in path imply that the novice should follow instructions with fewer "back-up" plans, yet those with alternative routes should be followed by the expert?