APPENDIX II

BACKGROUND

Workforce Analysis Model

Prior to 1991, INS used primarily historical workload to identify staffing levels and allocate new inspector positions at its POEs. However, this method was subject to influences both internal and external to INS that biased POE staffing. In 1991, INS awarded a contract to obtain a more objective method of determining staffing levels. The goal was to develop a mathematical staffing model, at the urging of Congress, that could be used to serve as "justification for additional resources." In response, INS awarded the development of the model to an outside contractor.

During the design of the model, the contractor and Headquarters Inspections conducted approximately 60 site visits at selected POEs in order to understand the staffing needs of the POEs. The initial contract was completed in 1992. The result was the design of WAM, a computerized mathematical model which projects the additional staff required at all types of POEs.

WAM contains features which simulate POE operations and results in both screen animations and written reports. There are three major components in WAM: WAMinput, WAMrun, and the General Purpose Simulation System (GPSS).

• WAMinput. The input and edit of POE data related to workload.

• WAMrun. The conversion of workload data into a form used to create the simulations.

• GPSS. Software programs that create the simulations and generates a screen animation or a written report.

The POEs inspection workload data may be entered directly into WAM, or through the larger INS Port of Entry Office Management System (POMS), of which WAM is a component. Once per year, each POE is required to input specific workload information relative to the port. During our audit, we found that 36 of 41 POEs (88 percent) sampled had computer equipment necessary to enter data into WAM.

In various versions of WAM's design, output reports can be generated using several groups of variables which alter the projections of the number of inspectors needed in total or on an hourly basis to process a given workload at a port. Examples of the various groupings of variables that generate output reports are:

• no growth or a percentage of growth in the workload;

• high, medium, low, or composite ¹⁰ weeks of seasonal workload data;

• local or standard inspection processing times; and

• required ¹¹ or optimized ¹² runs.

Another design feature in WAM was to project the number of additional inspectors needed at a POE on an hourly basis for each day of the week. One other feature was to limit the staffing projections of additional inspectors needed, for primary inspection work, to the maximum capacity of a port. ¹³

Since 1993, the WAM contractor's efforts have primarily been to train INS staff to enter data into WAM, to analyze the POEs' workload and staffing data submitted, and to summarize the POEs' workload and staffing data for Headquarters Inspections use. The WAM contractor performed the optimization runs for all ports and generated the WAM output reports in FYs 1995 and 1996. (See Statement on Management Controls, page 18.)

From FYs 1991 through 1995, INS has spent approximately $1.35 million to design, install, modify, and operate the model. The cost of the contractor's efforts for WAM was approximately $500,000 in FY 1995.

INS managers are beginning to use WAM output to justify budget requests for additional staff, as well as to respond to Congressional inquiries into port staffing levels.

Staffing and Quality Control Process

Headquarters Inspections begins the overall staffing process by selecting the highest, lowest, and median months of inspection traffic. The ports select the week within the target months and then generate the computerized WAM input data sets ¹⁴ using the workload tracking modules in the Port of Entry Office Management System (POMS). The POEs not using the tracking modules in POMS on a continual basis must enter the required data directly into WAM. A telephone Help Desk is available to answer questions and provide a "walk through" of the process with data entry personnel who encounter problems. In addition, POMS/WAM training teams travel regularly to POEs to educate users in the methodology for providing input into POMS and WAM. The training teams stress the need for generating accurate data.

Each data set received from the ports is reviewed by the WAM contractor to detect major data entry errors. For each air and land POE, the data sets are merged by the WAM contractor to produce a composite demand ¹⁵ which is then used to produce a baseline validation run. ¹⁶ This baseline run is then subjected to a series of quality control checks that compare the input and output with prior years data and the G-22.1 statistics.¹⁷   The baseline run is also looked at port-by-port to locate any obvious anomalies and detect any remaining errors in the data sets that may produce WAM run failures.

After a successful baseline validation run, the WAM contractor prepares an optimization ¹⁸ run that uses the composite demand with national standard inspection processing times and estimated growth. Headquarters Inspections determines the final growth percentage for use in the final optimization runs by comparing the POEs' estimates to the applicable industry estimates and historical trends. The growth optimization run is used for staffing projections purposes.

Headquarters Inspections prepares two Consolidation Staffing Requirements and Allocation Model (CRAM) spreadsheets that receive the data from the growth optimization runs. The spreadsheets aid in the overall quality control process by comparing the WAM results, ¹⁹ by port, to the authorized positions. The CRAM spreadsheets also adjust final staffing estimates by factoring in inspectors' indirect time (annual leave, sick leave, holidays, and training) and supervisory and administrative overhead. As part of the overall WAM process, trend analysis of previous years POE data is completed at Headquarters Inspections.

Other Headquarters Inspections quality control procedures include overtime audits that are performed on selected POEs by district and regional office staff. The overtime reports are reviewed to ensure that overtime funds for inspector staffing purposes were used appropriately and in accordance with the directives. The overtime audits include tests that compare shift schedules to workload patterns in an effort to control situations of excessive overtime.

After INS's authorization and appropriation bills pass the Congress and are signed by the President, Headquarters Inspections inputs the number of additional inspectors approved by Congress into CRAM. CRAM distributes the new positions to those ports with the greatest need. WAM's overall results support the distributions of the new inspector positions.

Headquarters Field Operations, the Regions, and District Office management review and resolve non-concurrence issues with the proposed allocations of additional inspectors. Eventually, final allocations (and funding) are distributed to the Regions. The District Offices are responsible for recruiting and hiring new inspectors.

Growth of Workload

According to INS, inspections were conducted at 241 locations in FY 1994. Additionally, INS statistics report the number of persons inspected at all POEs has grown from 457 million in FY 1990 to 509 million in FY 1994, or about 11 percent, as shown in the following table.

        Table 2

To process this massive workload, the total number of INS inspectors has grown approximately 16 percent from FY 1990 to FY 1994. The staffing at air and sea POEs has increased each year to keep up with the workload. However, at land border POEs, the number of inspectors has decreased 10 percent, as shown in the table below.

        Table 3

In FY 1994, the number of airport and seaport inspections represented 13 percent of the total inspections performed by INS. The remaining 87 percent (over 443 million) of all persons entering the United States applied for admission at land border POEs.

Pertinent Laws and User Fees

The Immigration and Nationality Act, as amended, requires all persons seeking entry into the United States present themselves for inspection. Immigration Inspectors determine their admissibility by inspecting required documents and interviewing the applicants. U.S. citizens are automatically admitted upon verification of U.S. citizenship. Aliens are admitted if they have complied with the requirements of the Immigration and Nationality Act.

The 1987 Appropriations Act established the Immigration User Fee account to fund immigration inspections. The law directs the Attorney General to charge and collect a user fee from each individual arriving in the United States aboard a commercial aircraft or vessel from foreign locations (other than Mexico, Canada, and the adjacent islands).

In the 1991 Appropriations Act, Congress removed the fee exemption for Mexico, Canada, and the adjacent islands for air travelers. The additional income was to be used to provide adequate staffing levels to ensure INS was in compliance with the 45-minute standard. This act also provided for the establishment of a pilot project under which a fee may be charged and collected for immigration document services at one or more land border POEs. The land border fee collection program was to be in operation by October 1995. Since our field work was conducted prior to the initiation of the fee for immigration document services at land border ports, we were unable to determine the extent of the effect additional funding may have had on additional land border inspector positions in the future.

Immigration User Fee Account receipts have increased substantially over the past several years as shown in the chart on the following page.

Figure 2

10 Composite weeks are a combination of the high, medium, and low seasonal weeks of workload.

11 Required runs display the number of inspectors currently assigned to the port, but will not project the number of additional inspectors needed in total or on an hourly basis.

12 Optimization runs project the number of additional inspectors that are needed in total or on an hourly basis to process a given workload.

13 As shown by Figure 5, Appendix III, WAM will not detect staff scheduling beyond port capacity.

14 Includes inspection workload, inspectors' shift schedules, remote assignments, POE inspection processing times, and basic port configuration information.

15 Composite demand is a WAM run which combines the high, median, and low weeks of seasonal workload data entered into WAM.

16 A baseline validation run is a computer run to search for conditions, such as missing data, that will cause the model to fail during later optimization runs.

17 Workload statistics generated by the ports of entry.

18 An optimization run will generate a projection of the number of additional inspectors needed over and above the existing inspectors at a port.

19 WAM results are the output projections of the additional inspectors needed, using standard inspection processing times, plus the existing staff at a port of entry. The WAM results are stated in terms of permanent full-time inspectors and other-than-full-time inspectors.

#####