Introduction to Transaction Processing and Accounting Information Systems

Introduction to Financial Transactions and the Three Transaction Cycles

A financial transaction is defined as an economic event that affects the assets and equities of the firm, is reflected in its accounts, and is measured in monetary terms. In an Accounting Information System (AIS), similar types of transactions are grouped together into three primary transaction cycles: the expenditure cycle, the conversion cycle, and the revenue cycle. The relationship between these cycles illustrates how data and resources flow through an organization to maintain its financial health.

Each cycle consists of two primary subsystems. In the expenditure cycle, there is typically a time lag between two components due to credit relations with suppliers. These components are the physical component, which involves the acquisition of goods, and the financial component, which involves cash disbursements to the supplier. The conversion cycle consists of the production system and the cost accounting system. The production system involves the planning, scheduling, and control of the physical product through the manufacturing process, while the cost accounting system monitors the flow of cost information related to production. The revenue cycle also features a time lag between its two components due to credit relations with customers. These components are the physical component, which involves sales order processing, and the financial component, which involves cash receipts.

Manual System Accounting Records and Document Types

Manual systems rely on various types of accounting records to capture and formalize data. Source documents are used to formalize transaction data needed for processing. Product documents are the tangible result of transaction processing. Turnaround documents are a unique category where a product document of one system becomes a source document for another system later in the process. Journals provide a record of chronological entries. There are special journals used for specific classes of transactions that occur in high frequency, and a general journal used for nonrecurring, infrequent, and dissimilar transactions.

Ledgers function as a book of financial accounts. The general ledger shows activity for each account listed on the chart of accounts, providing a summarized view of the firm's financial status. In contrast, a subsidiary ledger shows activity by detail for each account type, allowing for specific tracking of individual customers or vendors. The flow of information typically travels from an economic event into source documents, then into journals, and finally into the general ledger and subsidiary ledgers.

Accounting Records and Audit Trails in Digital Systems

In computer-based systems, the audit trail is less observable than in traditional manual systems. This is because the data entry and computer programs constitute the physical trail, and the data is stored in magnetic files rather than paper books. Accountants must be able to trace transactions in both directions: from a source document to the financial statements and from the financial statements back to the source document. Sampling and confirmation are two common techniques used to verify these records.

Specific steps to verify accounts receivable ($AR$) in a digital system include comparing the $AR$ balance in the balance sheet with the master file $AR$ control account balance. The auditor then reconciles the $AR$ control figure with the $AR$ subsidiary account total. A sample of update entries made to accounts in the $AR$ subsidiary ledger is selected and traced back to transactions in the sales journal, which is often an archive file. From these journal entries, the auditor identifies physical source documents—such as sales orders, deposit slips, remittance advice, or shipping notices—that can be pulled from files and verified. If necessary, these documents are confirmed by contacting customers directly.

Data Storage Models and Computer File Types

There are two primary models for data storage in AIS: the flat-file model and the database model. The flat-file model is often plagued by data redundancy problems. These include excessive storage costs for maintaining multiple paper or magnetic copies, data updating issues where changes must be performed multiple times, and problems with the currency of information if all affected files are not updated. Additional issues include task-data dependency, where a user cannot obtain additional information as needs change, and data integration problems where separate files are difficult to combine across different users.

Computer systems utilize four specific types of files. Master files generally contain account data, such as the general ledger and subsidiary files. Transaction files are temporary files containing transactions that have occurred since the last update. Reference files contain relatively constant information used in processing, such as tax tables or customer addresses. Archive files contain past transactions maintained specifically for reference and audit purposes.

Documentation Techniques for Systems and Logic

Documentation in a digital environment is essential and is achieved through several common techniques. The Entity Relationship Diagram ($ERD$) represents the relationship between entities in a system using cardinalities to represent numerical mapping. These mappings include one-to-one ($1:1$), one-to-many ($1:M$), and many-to-many ($M:M$). For example, a salesperson may be assigned to a car ($1:1$), or a vendor may supply many inventory items ($M:M$).

Data Flow Diagrams ($DFD$) use symbols to represent processes, data sources, data flows, and entities. Unlike other diagrams, $DFD$s represent the logical elements of the system rather than the physical system. System flowcharts illustrate the relationship among processes and the documents that flow between them. They provide more detail than $DFD$s and clearly depict the separation of functions. These flowcharts represent the relationship between key elements—input sources, programs, and output products—and depict the media used, such as paper, magnetic tape, magnetic disks, or terminals.

Program flowcharts are used to illustrate the specific logic within computer programs. Symbols used in program flowcharts include a rectangle for a logical process, a diamond for a decision, an oval for terminal start or end operations, and a parallelogram for input/output operations. Each symbol is connected by flowlines indicating the flow of the logical process.

Transaction Processing Models: Batch versus Real-Time

Batch processing involves grouping similar transactions accumulated over time and processing them together. For batch processing to be appropriate, the transactions must be independent of one another during the accumulation period. There is always a time lag between the economic event and the processing. Steps in batch processing using sequential files include: the Keystroke (transcribing documents to magnetic tape), the Edit Run (identifying clerical errors), the Sort Run (rearranging data by a primary key to match the master file), the Update Run (changing master file values), and the Backup Procedure (creating a "son" file from the "father" file).

Real-time systems process transactions individually at the moment the economic event occurs. There is no time lag in these systems. While real-time systems generally require greater resources and dedicated processing capacity, the cost differentials between batch and real-time are decreasing. Many AIS still use batch processing because certain tasks, like recording high-volume cash receipt checks, are characterized by independent transactions that can be processed during off-peak computer times using real-time data collection.

Data Coding Schemes in AIS

Data coding is used to concisely represent complex information, provide accountability, identify unique transactions, and support the audit function. Sequential codes represent items in numerical order and are often used to prenumber source documents to identify out-of-sequence items. However, they carry arbitrary information and are hard to change. Block codes represent whole classes by assigning ranges, such as for a chart of accounts. This allows for easy insertion of new codes but remains somewhat arbitrary.

Group codes represent complex items using fields with specific meanings. For example, a code like $04-09-476214-99$ might represent Store Number $04$, Department Number $09$, Item Number $476214$, and Salesperson $99$. Alphabetic codes can represent up to $26$ variations per field and can be used in block or group structures. Mnemonic codes use abbreviations or acronyms to be informative rather than arbitrary, such as using "NY" for New York. While informative, mnemonic codes have limited usability and availability depending on the dataset.