The primary objective of this study is to examine the performance of order-
based dispatching rules in a general job shop, where the environmental fact
ors are shop utilization and due date tightness. An order is defined as a c
ollection of jobs that are shipped as a group-an order-to the customer, onl
y on completion of the last job of the order. We specifically compare dispa
tching rules from past job-based studies to some rules adapted to encompass
order characteristics. Standard flow time and tardiness measures are used,
but in addition, we introduce measures that combine average performance wi
th variation in an attempt to capture the performance of a majority of the
orders processed in the shop. Of the 16 dispatching rules tested, our resul
ts show that four of the simple rules dominate the others. We also found th
at order-based rules perform better than their job-based counterparts. The
study makes use of multivariate statistical analysis, in addition to the us
ual univariate tests, which can provide additional insight to managers usin
g multiple criteria in their decision process.