In-patient bed requirement (or number of beds required to cope with th
e workload) is of prime importance for hospital management because of
its direct implication for manpower requirement and therefore the cost
s. In-patient beds could be divided into emergency beds and elective b
eds, where the requirement for emergency beds will be affected by both
patient length of stay distributions and the random patterns in which
patients arrive. In this paper, the author extends the early results
by Pike et al.(1963, Br. J. Prev. Sec. Med. 17. 172) so that day-of-we
ek effect is introduced into the queuing model to evaluate the require
ment of emergency beds. It is demonstrated that emergency bed occupanc
y on any specific day of the week follows a Poisson distribution.