Credit Scoring

The act of credit scoring entails evaluating the likelihood that a credit applicant will default according to a statistical or nonlinear model derived from past clients’ socio-demographic data (e.g. income, expenses, employment status, etc) and their resulting credit behaviour.  Basically, the historic clients’ dataset is data-mined to unearth certain patterns that are predictive of creditworthiness.  Once these patterns have been discovered and the model built, it is then used to classify new applicants as either “good” or “bad” for credit.

​

ECL Provisioning

Expected credit loss (ECL) forecasting involves the act of evaluating the likelihood that a loan will default (i.e., 90 days past due) according to a statistical or nonlinear model derived from past clients’ socio-demographic data and credit behaviour, and then the multiplication of this estimate of the probability of default (PD) by the expected loss given default (LGD) and exposure at default (EAD).  Thus, the basic methodology of the ECL forecasting model developed is based on the estimation of these key parameters which are in turn derived from the Financial Institutions’ historical data and available macroeconomic data.

​

Predictive Maintenance

When predictive maintenance is working effectively as a maintenance strategy, maintenance is only performed on machines when it is required. That is, just before failure is likely to occur.

This brings several cost savings by minimising i) the time the equipment is being maintained, ii) the production hours lost to maintenance, and iii) the cost of spare parts and supplies.

​