ABSTRACT
Research has shown that construction contractors face challenges when attempting to manage their present and future financial requirements, via cash flow (CF) forecasting and/or the use of cash flow forecasting (CFF) models. Consideration for risk factors impacting on CF forecasts/CFF models has also been identified as a key issue affecting contractors' application of CFF models. The aim of the research is to develop models for predicting the impact of risk factors on CF forecasts predicted by contractors in Nigeria. A list of risk factors that impact on CF forecasts were identified through the review of existing literature on cash flow management, cash flow forecasting, risks in cash flow forecasts and factors contributing to project delays. The identified risk factors were then investigated in relation to the likelihood of occurrence and the impact on CF forecasts, if they occur. Data was collected via a questionnaire survey of contractors operating in the Nigerian construction industry. Two (2) sets of questionnaire surveys were conducted. The first questionnaire survey focused on determining the likelihood of occurrence/impact of forty (40) and thirty-one (31) risk factors impacting on cash-in and cash-out forecasts respectively. Contractors were asked to rank on a scale of 0 to 5 the likelihood of occurrence/impact of the risk factors. The responses obtained were subjected to analysis with the use of IBM SPSS (version 21) software. The mean, standard error and standard deviation were computed. The computed means were used to rank the likelihood of occurence/impact of the factors in decending order. Seventeen risk (17) factors were found to significantly impact on "cash-in" forecasts- "delay in receiving retention", "accuracy of estimates", "change in government officials", among others. While eighteen risk (18) factors were found to significantly impact on "cash-out" forecasts- "increased duration of the project", "change in currency exchange rates", "high cost of materials", among others. The second survey focussed on sixteen risk (16) factors whose likelihood of occurence and impact on cash-out forecasts were found to be significant. The survey sought respondents to estimate the percentage variations between actual and forecasted "cash-out" forecasts owing to the occurence of the significant risk factors. The responses of the second survey were fed as input to IBMSPSS neural network software and used to develop neural network models for predicting the impact of the significant risk factors on "cash-out" forecasts. The models developed showed that construction contractors' cash-out forecasts vary with the actual expenditure by +20%, +25% and +25% at the 30%, 50%, and 70% completion stages respectively. Validation of the models shows a 77%, 69%, and 67% accuracy at 30%, 50%, and 70% completion stages respectively. The findings from the research imply that several risk factors have different degree of occurrence on cash-in and cash-out forecasts in, and different degree of impacts on cash-out forecasts in the Nigerian construction industry. The impact on cash-in was not modelled as literature had suggested that modelling the impacts was not a true representation of reality. Construction contractors practicing in the Nigerian construction industry should expect positive variations to their "cash-out" flow forecasts during the execution of projects. Major recommendations include; contractors should carefully consider the seventeen (17) and Eighteen (18) highest occurring and high impact risk factors affecting "cash-in" respectively, before embarking on any construction project. The Eighteen (18) and Sixteen (16) highest occurring and high impact risk factors affecting cash-out respectively, should be considered before embarking on any construction project.