Integrating Linear Regression and Multi-Criteria Decision Making for Assessing Financial Statement Risks in Manufacturing Firms

TL;DR

Manufacturing firms face interdependent economic, operational, and managerial factors that traditional discounting methods fail to capture. The paper proposes an integrated framework that translates costs and benefits into present-value terms using compound-interest discounting and then uses linear regression with Average Weight Scores (AWS) to quantify each criterion's marginal contribution to discounted economic performance. It formalizes a regression model $\bar{R}_i = \beta_0 + \beta_1 AWS_{2,i} + \beta_2 AWS_{3,i} + \varepsilon_i$ to map multicriteria weights to discounted outcomes, noting a moderate fit around $R^2 \approx 0.55$ and providing interpretable rankings of criterion importance. The approach offers a transparent, data-driven method to guide long-term, value-aligned decision-making for control system efficiency in manufacturing, with future work suggesting nonlinear and dynamic extensions and industry-specific AWS calibration. The framework bridges theoretical discounting with empirical, interpretable analysis, enhancing robustness and actionable insights in financial performance evaluation.

Abstract

Evaluating the financial performance of manufacturing firms requires consideration of both the time value of money and the relative importance of multiple decision criteria. Conventional approaches relying solely on deterministic discounting often fail to account for interactions among economic, operational, and managerial factors. This study proposes an integrated framework that combines time-discounted economic analysis with linear regression to evaluate control system efficiency. A theoretical discounting model is first developed to convert costs and benefits occurring at different times into present-value terms using compound interest functions. The model accommodates one-time expenditures, time-proportional costs, and complex cost structures arising during system development and commissioning. To empirically assess how discounted economic performance is influenced by multiple criteria, linear regression serves as the approximation method.

Figures (2)

• Figure 1: Linear regression example among Table \ref{['tabeqq1']}.
• Figure 2: Linear regression analysis between secondary criteria weights ($AWS_2$) and overall weights ($AWS_1$)