Quality control charts have been widely utilized in manufacturing and service industries since the day they were introduced. The Cumulative Sum (CUSUM) chart is one of the most common charts utilized for monitoring both attribute and variable data. It is noticed from the literature, that many efforts have been conducted to optimize and enhance the performance of the CUSUM chart. This research proposes a modified CUSUM (wCUSUM) scheme for monitoring attributes, in which the difference between the actual and in-control numbers of nonconforming items is raised to the power of an exponent w to improve the efficiency of the chart. The operation of the developed wCUSUM is compared to that of the semi-optimal CUSUM chart established by Wu et al. (2008) under different cases using the Average Number of Defectives (AND) as an objective function. The wCUSUM scheme managed to outperform the other CUSUM scheme against a wide range of shifts in fraction nonconforming (p). A detailed comparison is conducted to assess the performance of the wCUSUM and semi-optimal CUSUM charts, while assuming the p shift to follow both uniform and Rayleigh distributions. On average, the proposed wCUSUM is found to outperform the semi-optimal CUSUM chart by 21% and 60% using uniform and Rayleigh distributions of p shift, respectively. Sensitivity analysis is performed where the effect of the design parameters (the false alarm rate τ, in-control fraction nonconforming p0, maximum shift δmax, and sample size n) on the response, AND, is studied. A case study is conducted in the solar panel manufacturing industry to demonstrate the design and application of the wCUSUM. In this case study, the wCUSUM chart is able to enhance the detection effectiveness by 65% in terms of AND, compared with the semi-optimal CUSUM counterpart