...threw up some interesting explanations:

https://titan.uio.no/sites/default/files/unmanaged/smith_et_...

(but you'll have to search through the PDF file)

https://books.google.fr/books?id=xAbTBwAAQBAJ&pg=PA35&lpg=PA...

This second ref. gives a better explanation

And here's another:

https://books.google.fr/books?id=eCTD6PlO3Y8C&pg=PA99&lpg=PA...

It seems that any particular 'reading day' is the day on which the data is read.

2.2.1. Additive models
Adams and Vodicka (1987), in their study of forecasting at Qantas Airlines, developed short-term forecasts for just 1 week before departure. They used fairly simple methods which relied on subjective marketing estimates and simple averages of segment class reservations. L’Heureux (1986) discussed the classical ‘pickup’ (pickup is defined as the number of reservations picked up from a given point in time to a different point in time over the booking process) method and the advanced pickup method in an airline context. The classical pickup method determines the average (or weighted average) of reservations picked up between different reading days (e.g., between 120 days out and 90 days out) for departed flights (e.g., CP 121 CalgaryMontreal) for a particular day of week to forecast the future pickup between the same reading days for the same flight number on the same day of week in the future. The advanced pickup method is similar to the classical pickup method, with the extension that it includes relevant data from all flights, even those that have not yet departed.

http://scholarship.sha.cornell.edu/cgi/viewcontent.cgi?artic...


--------------------------------------------------
Note added at 2 hrs (2016-08-06 21:42:51 GMT)
--------------------------------------------------

'Advanced booking models forecast the number of booked rooms on a particular arrival day on the basis of the number of booked rooms on a previous day (called "reading day") and the "pick-up" of rooms...'

https://books.google.co.uk/books?id=ZSrJAwAAQBAJ&pg=PA83&lpg...

--------------------------------------------------
Note added at 2 hrs (2016-08-06 21:48:15 GMT)
--------------------------------------------------

Advanced booking models forecast the number of booked rooms on particular arrival day on the basis of the number of booked rooms on a previous day (called “reading day”) and the pick up of rooms between the reading day and the arrival day. Weatherford and Kimes (2003, p. 403) divide advanced booking models into additive and multiplicative models. Authors explain that additive models assume that the number of reservations on hand at a particular day before arrival is independent of the total number of rooms sold. In these models the number of booked rooms on the reading day is added to the average historical pick up between the reading and the arrival day. On the other hand, multiplicative models assume that the number of reservations yet to come is dependent on the current number of reservations available (Weatherford & Kimes, 2003, p. 403).