Among the many clichés that one hears in today’s harried business environment is the phrase “time is money.” While a cliché by its nature is worn out, this one is quite descriptive of the current demands on Information Systems. Traditional periodic mode systems that provide information primarily through periodic reports that are hours, days, or weeks out of date can put an organization’s decision makers at a disadvantage if its competitors are using up-to-date information to make the same decisions (e.g., recall the importance placed on timeliness and relevance in Chapter 1). The pressures for timely information flows coupled with significant advances in available information technologies led to a rapid migration towards online real-time systems. Online real-time (OLRT) systems gather business event data at the time of occurrence, update the master data almost instantaneously, and provide the results arising from the business event within a very short time—i.e., in real-time. OLRT systems complete all stages of business event data processing in immediate mode. Immediatemode is the data processing mode in which there is little or no delay between any two data processing steps (as opposed to periodic mode, in which there is a significant delay between two or more data processing steps).
Explain the relationship between online real-time (OLRT) and immediate mode processing.
OLRT systems typically require three basic subprocesses to be completed before an event is converted into information that can be used by decision makers. Figure 4.3 illustrates each of these subprocesses.
- Business event occurrence and recording of event data: At the time of the business event, related data are entered directly into the system. Source documents are almost never used, as they significantly slow the process and remove some of the advantages of nonredundant data entry. Notice that the data entry process where the sale is entered into the system is the same as in Figure 4.2 (other than the absence of an audit file). This process is consistent with the use of online transactionentry (OLTE) for OLRT systems.
- Update master data: Each business event entered into the system is processed individually and any calculations and summarizations completed. This information is then used to update the master data. Note in Figure 4.3 that the processing is now being done on-site where the sales event data are entered.2 Because each business event is processed independently and immediately, the master data at any given time will be within seconds of being up to date. When your books and CDs store is entering your information into the computer, it may be using an OLRT system if it is important to the store to know whether a given book or CD title is in stock at a given time—perhaps to answer a customer’s question.
Generate reports and support queries: It is neither practical nor desirable that reports be generated after each business event is recorded and master data have been updated. Typically, applicable reports are generated by the system on a periodic basis. At the same time, however, these reports are usually instantaneously available through access to the system on an as-needed basis, as demonstrated in Figure 4.3 with the communications links to the sales and inventory managers. One of the main advantages provided by many OLRT systems is an ability to check the current status of master data items at any given time. In the books and CDs store, it would allow the sales staff to check quickly whether a given book or CD is in stock. In many cases, rather than using pre-specified reports that may not necessarily provide information that decision makers need, these Information Systems users use a query language (as discussed in Chapter 3) to create unique reports dynamically that provide the one-time information they need to make key decisions. For instance, the store manager may want to run a report on the inventory stock for the top-ten selling CDs and books.
List and describe the three basic subprocesses completed in processing business event data using online real-time processing.
It was noted previously that OLTE systems are also increasingly used with systems that primarily use the periodic mode. While the data entry performed in all OLTE systems is essentially the same, the mode of processing may vary. While a pure periodic mode system still processes business event data in batches, an OLRT system using OLTE processes each recorded business event in real time. In a real-time system, business event data cannot be aggregated on a local computer to be transferred later to the data processing center. Rather, each business event must be communicated for processing at the time the event occurs. This results in a more expensive approach to OLTE. In essence, rather than creating a temporary electronic communications connection to download data to the data processing center, an OLRT system generally requires a continuous electronic communication connection, usually necessitating the use of some form of network. This arrangement will be addressed later in this chapter.
It should be noted here that automated systems that model manual systems and OLRT systems are the two extremes in business event data processing. The systems that mimic manual systems are what we might term pure periodic mode systems in that there is a delay between every step of the processing. On the other hand, OLRT systems represent pure immediate mode systems in that there is little or no delay between any steps in the processing. We note these as the extremes because many systems lie somewhere between these two extremes, exhibiting a mix of periodic and immediate mode processes at various stages. For example, OLTE used with batch processing results in an immediate mode approach for combining the business event occurrence and record event data steps, while periodic mode processing might be used for the remainder of the steps.