July 17th, 2006

Australian Cargo

Posted by Michael Krigsman @ 9:23 pm

Categories: Government projects, Project failures, Project management

Tags: Michael Krigsman

David Jacobson’s blog describes a report prepared by Booz Allen Hamilton, analyzing problems in the Australian Integrated Cargo System. The system is intended to track goods coming into Australia by either air or sea. A variety of implementation problems with system resulted in traffic congestion and escalating costs, and the report was an attempt to figure out what happened.

According to the report, the complexity of the project, ”combined to provide a challenging environment with multiple lessons for future implementations.” The implementation involved multiple government agencies, a wide-range of external companies, and third-party software suppliers, adding to the difficulty of the project.

David summarizes key problem sources:

• Inadequate end-to-end testing of the system – testing with live data would have highlighted many of the issues that manifested when the system went live
• Quality problems with third-party software – many users experienced problems that were not ICS-related but rather were problems in the software they use to communicate with Customs. Customs has no means of checking their readiness.
• A lack of staged implementation – the “big bang” implementation provided no realistic fall-back when problems existed and no way of ensuring that the total end-to-end importation supply chain process worked correctly before widespread implementation;
• Inadequate training in the new system – many organisations were not able to train in the combined new environment of their third-party software and the ICS. This meant that these organisations were not sufficiently prepared and had not adapted their own internal processes to the requirements of the new system.

The report specifically cites a variety of IT governance issues, including training, program management, usability testing, and so on. Sounds like a poorly managed and executed Frankenstein-style project.

The full report can be read here (.pdf file). The report is well-written and definitely worth a read if you are interested in large-scale implementation management.

—–