According to the Kroll report, one area in supplier relations where IT can help discover fraud is in analyzing billing and payment processes. Sullivan notes that software applications, which mine a company's databases and create exception reports, can detect unusual payment patterns. "When controls around payment matching and approvals are weak, service providers will learn that a company does not notice when they over bill, double bill, ghost bill, or bill for the wrong service," Sullivan notes in the report. "Falsified invoices, however, rarely follow the same patterns as those from honest suppliers."

Sullivan lists four red flags as possible indicators of fraudulent activity: one, a dramatic increase in payments to one vendor; two, a high number of transactions under audit thresholds; three, consecutive invoice numbers or multiple invoices on the same day; and four, payments highlighted by exceptions to what's referred to as Benford's law (which holds that numbers that occur naturally in business typically follow some basic patterns.) "Perpetrators of fraud rarely have the knowledge or capability to fake these," he notes.

Due to the complexity of a company's logistics needs, there is also a need to use software packages to detect adverse behavior, notes Eduardo Gomide, a managing director in Kroll's financial advisory services, in the report. Gomide defines logistics as the procurement, movement and storage of materials, parts and finished inventory throughout a company and its market channels.

"The vulnerabilities are considerable in part because of the field's complexity, involving as it does different parties and a huge flow of information that can be difficult to analyze, predict, or even measure," Gomide notes.

One approach to minimize companies' exposure to logistics fraud is to mine their own databases using commercially available statistical software that can present critical graphical and numerical analyses.

"The identification of patterns, outliers and other anomalous activities can then generate and trigger alerts, as well as identify leads for investigation and audit," Gomide states. "The approach sounds complex, but...in essence, this simply involves the application of common sense to large volumes of existing data from accounting, stock control or ERP systems."

In the end, "whether red flags are present or not, well-designed internal controls and functional information systems make these frauds difficult to perpetrate," Sullivan says in the report. "We have yet to see fraud seep into a logistics department where strong controls and processes are present."