In-use emissions from vehicles using heavy-duty diesel engines can be significantly higher than the levels obtained during engine certification. These higher levels may be caused by a combination of degradation of engine components, poor engine maintenance, degradation or failure of emissions after-treatment devices, and engine and emissions system tampering. A direct comparison of in-use vehicle emissions with engine certification levels, however, is not possible without removing an engine from the vehicle in order to perform engine dynamometer emissions testing. The goal of this research was to develop a chassis test procedure that mimics the engine performance, and as such the expected emissions levels, from the engine certification emissions test prescribed in the U.S. Code of Federal Regulations. Emissions measurements were taken from two engines during testing on an engine dynamometer using the transient heavy-duty Federal Test Procedure (FTP). Additionally, each engine was installed in an appropriate vehicle, and emissions measurements were taken using a chassis dynamometer while employing a vehicle driving schedule IMPLICATIONS At present, it is difficult to determine how engines degrade in real use because they are certified for emissions independently of the vehicle, and are seldom removed for retesting. This research has provided a methodology for testing a whole vehicle, with a single gear selected, in a fashion that causes the engine to follow the engine test certification procedure as closely as possible. The results show a strong correlation between engine-and chassismeasured results for NO x , so that engines with high NO x emissions can be identified without removing the engine from the vehicle if this approach is used. Deterioration of NO x emissions can therefore be quantified for a fleet. Data show that using this approach to quantify PM emission is less precise, and that the chassis-based approach may be limited to finding only "super-emitters" if applied to PM compliance.intended to match closely the instantaneous torque and speed schedule of the engine FTP.Engine and chassis testing was performed with the engines in stock (unmodified) condition as well as in several modes to simulate either tampered or poorly maintained conditions. The use of a chassis test as a predictive tool for determining whether an engine in a vehicle would pass the engine certification test has proven to be worthwhile. Analysis of the data shows that identification of chassis-mounted engines with NO x emissions above certification levels is possible by employing engine-specific correction factors. In the case of PM emissions, significant data scatter allowed only the identification of gross PM emitters. Engine tampering and poor maintenance can raise PM and NO x emissions, and these increases can be correctly identified by a chassis test. Analysis of chassis and engine CO and HC emissions did not reveal a strong enough correlation to warrant the use of the chassis test for emissions screening of these two po...