App gives Customs a gauge | GCN

Customs’ Michael Raithel and
Sandy Koncir collaborated on the CMAS intranet application, which has been nominated for a
Computerworld Smithsonian Award.


Using the Customs Measurement Analysis System scheduled to come online this month,
Customs Service officials for the first time will see how well they are doing their jobs
compared with peers at other ports and Customs Management Centers.


The accountability measures are supposed to help managers do a better job of
supervising the agency, which oversees all persons and goods entering the United States.


Customs measures include the number of narcotics seizures from cargo, trade compliance
rates and percentage of air travelers who clear customs within five minutes.


The port-to-port comparison number is a favorite among Customs officials, according to
senior systems analyst Michael Raithel.


In the new culture of accountability, agency executives agreed on 18 measures for
reporting on its performance to Congress, as all agencies are required to do under the
1993 Government Performance and Results Act.


Most agencies are still coming to terms with the requirements. Customs is ahead of them
in automating the extraction of GPRA measures that it reports every six months to the
assistant commissioner of field operations.


The agency used intranet tools to create a fully automated system that calculates
management results uniformly.


“It’s been exciting for me, an old mainframer—well, longtime
mainframer—to play with the tools,” said Raithel, an experienced SAS programmer.


Raithel wrote the Customs Measurement Analysis System (CMAS) using SAS Release 6.12 and
SAS IntrNet 1.1 for Microsoft Windows NT 4.0. Both fourth-generation language tools come
from SAS Institute Inc. of Cary, N.C.


Because CMAS uniformly calculates GPRA measures, “somebody at headquarters will
get the same numbers as somebody in New York or in San Francisco. The measures are
repeatable and verifiable,” Raithel said


All 18 CMAS measures are based on data from Customs’ Office of Management
Reporting data warehouse [GCN, Oct. 19, Page 1].


The data undergoes a rigorous registration process. Each data element in the warehouse
has a data owner “to make sure the element is in tip-top shape, or the best shape we
can get it into,” Raithel said


CMAS has its own data dictionary. If people are unhappy with a GPRA measure they see in
CMAS, Raithel said, they can go back into the data warehouse to find the elements that
make up the measure, “click and find the enumerator and the denominator, do the
calculation themselves and verify the measure.”


The intranet application lets Customs managers look at GPRA measures every month within
two hours of the agency refreshing its OMR warehouse of management data.


The warehouse stores 320 data elements extracted from Customs’ CA-Datacom
mainframe database management system from Computer Associates International Inc.


The intranet application builds most of its pages dynamically. Raithel said he coded no
more than 20 Hypertext Markup Language pages.


SAS IntrNet, he said, “crunches the data and then builds reports in HTML.”


From the CMAS home page, managers are no more than four clicks away from getting the
reports they request, Raithel said.


CMAS executes several programs that use the SAS Access to Open Database Connectivity
interface to extract data elements from the OMR Microsoft Access database and load them
into a SAS data mart.


The 6M data mart runs under Windows NT 4.0 on a uniprocessor Dell Computer Corp.
PowerEdge 2200 application server on a Novell NetWare LAN 4.1. The data mart server has
130M of RAM and a 17G disk array.


To get the same kind of GPRA data as they did in the past, Customs managers had to
extract the data elements from the OMR warehouse every three to six months and do their
own calculations in a Microsoft Excel spreadsheet or Access database.


“Staff members had to become data processors,” Raithel said. “CMAS takes
a large burden off their backs.”


Managers can calculate GPRA measures for one month or a range of months. Or they can
compare the measures for two distinct months.


If they put a new management practice in place in September, they can compare the
month’s data to November data to see if the practice produced any relevant change in
the GPRA measure, Raithel said.


The Customs commissioner offered his own suggestions for expanding CMAS when he saw the
application for the first time a few months ago, Raithel said.


Because of desktop security concerns, Customs has moved more slowly than expected in
getting CMAS out to users, Raithel said.


Security personnel approved the application only for desktop systems running Microsoft
Windows NT 4.0 and Internet Explorer 3.02 or later versions.


Customs is in the midst of a large-scale deployment of Windows NT Workstation, but many
high-level managers who will use CMAS do not yet have NT on their desktops.


To speed deployment, Raithel rewrote the entire application to run on Customs’
secure IBM Corp. 9672/RY5 mainframe under OS/390 3.0.


He used SAS screen design programs and SAS 6.09E for IBM mainframes.


“It won’t be pretty, but at least they’ll have it,” Raithel said.


The mainframe version performs the same basic functions as the Web version but without
the niceties of cut-and-paste, local printing and report distribution via e-mail. “As
sweet as it is to roll CMAS out on the mainframe, there are an awful lot of positives on
the desktop that just can’t be beat,” Raithel said.


Customs completed the intranet application with a relatively modest investment in
software—less than $15,000—plus the salaries of programmer Raithel and business
analyst Sandy Koncir.


Customs’ GPRA application has been nominated for a Computerworld Smithsonian
Award, which means a description of the application will become part of the Smithsonian
Institution’s permanent research collection of socially significant technology
applications, officials said. 

inside gcn

  • pollution (Shutterstock.com)

    Machine learning improves contamination monitoring

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above