Influence of network structure on schedule performance – Extending criticality index to capture ripple effect of delays

Abstract

Many construction projects suffer delays. This situation is exacerbated in larger projects with their more complex schedules, because of a 'ripple effect'-the phenomenon that relatively small delays of activities may not just locally affect direct successors in a network schedule, but spread, accumulate, and globally impact its project finish date negatively. This research therefore studies the occurrence and behavior of said ripple effect by examining dependency structures of network schedules systematically. Its contribution to the body of knowledge is threefold: First, it develops a methodology that multiplies the elements from the correlation coefficient matrix with the reachability matrix to measure said ripple effect. It plots the cumulative product matrix in three dimensions to visualize the relative potential ripple effect of individual activities. Second, it conducts a sensitivity analysis by experimenting with schedule structures that range in complexity from sequential to parallel. A constant relation is found that the sum of the product of elements in the noncumulative productive matrix and its corresponding criticality index always equals one. Cruciality is defined. Third, it substantiates its results by simulating a benchmark schedule and derives practical suggestions for delay avoidance and delay analysis that considers how the network structure determines the behavior of a schedule.