Information Supply Chain: Operational Systems are Raw Inputs

BASICS: The Information Supply Chain > Raw Inputs (Operational Systems)

ISC Components:

The information supply chain begins with Operational systems, also known as “Transactional” or “OLTP” (online transaction processing), or even “ERP” (enterprise resource planning) – the main difference between ERP and the other terms being that the ERP system is more comprehensive; it integrates multiple areas of transaction processing together into one big system. From a DW / BI perspective, there’s little distinction, because transactional systems, however comprehensive, nevertheless all represent the same thing: they are the wellhead of enterprise data – the source of raw ingredients upon which the DW / BI system operates.

The core goal of operational systems is process enablement. These systems exist to make a business process run efficiently and smoothly, and with as little human administrative effort as possible to keep the process running. Typically within a business there will be a separate operational system for each discrete business process or function, or if an ERP is employed, a separate module (separate yet integrated to the other ERP modules) for each process or functional area. Some examples of different operational systems are depicted in the diagram above.

Operational systems have several distinguishing characteristics. From the perspective of user base, they tend to be used more by the rank-and-file employees of the company who have responsibilities for data entry and monitoring individual transactions, products, suppliers, and customers. This is not the case 100% of the time, but it is generally the case.

From the perspective of system design, operational systems are geared towards data entry, and so they are built for easy and speedy data capture. On the back end, they are designed to store that data in as compact and efficient a manner as possible, so as to keep the system small for performance reasons, but also to ensure the integrity of the data by means of normalization rules that prevent duplicate and ambiguous records from coming into existence that refer to the same real-world “thing” (customer, vendor, order, etc.).

The unit of work of an operational system is the individual transaction. That is, its job is to make sure the order gets taken, shipped, and invoiced, that the cash register dings, that the employee paycheck is cut, that the purchase order gets paid, etc. The operational system performs these tasks, over and over, in repetitive iterations. Once an iteration of work (representing a single business transaction) is complete and closed, the data the system captured for that transaction quickly stales. It’s really only useful to the system up thru the point that the transaction is completed. Thereafter, neither the operational system nor its users tend to pay any more attention to that record. Consequently, for performance reasons, and given the short useful life of transactional data to the operational system, a data archival plan is usually put in place to push aging data out of the system into secondary storage to keep the system small and well-tuned.

Some examples of prominent players in the Operational or ERP space include SAP’s ERP solution, Oracle Financials, JD Edwards, Peoplesoft, Microsoft Dynamics, and a host of other small mid and small market software vendors who offer complete ERP packages or point solutions targeted to specific process or functional areas.

The main point of Operational Systems, however, from the perspective of the information supply chain, is that they represent the point at which data is entered, and therefore the source of raw ingredients upon which the rest of the chain depends.