AUTHOR COPY Journal of the Operational Research Society (2010) 61, 1022--1031  2010 Operational Research Society Ltd. All rights reserved. 0160-5682/10 www.palgrave-journals.com/jors/ The newsvendor problem with unknown distribution U Benzion 1 , Y Cohen 2 and T Shavit 3 ∗ 1 Department of Economics, Ben-Gurion University of the Negev, Israel; 2 Department of Management and Economics, The Open University of Israel, Israel; and 3 School of Business, College of Management, Israel Newsvendor theory assumes that the decision-maker faces a known distribution. But in real-life situations, demand distribution is not always known. In the experimental study which this paper presents, half of the participants assuming the newsvendor role were unaware of the underlying demand distribution, while the other half knew the demand distribution. Participants had to decide how many papers to order each day (for 100 days). The experimental findings indicate that subjects who know the demand distribution behave differently to those who do not. However, interestingly enough, knowing the demand distribution does not necessarily lead the subject closer to the optimal solution or improve profits. It was found that supply surplus at a certain period strongly affects the order quantity towards the following period, despite the knowledge of the demand distribution. Journal of the Operational Research Society (2010) 61, 1022 – 1031. doi:10.1057/jors.2009.56 Published online 20 May 2009 Keywords: behavioural operation; newsvendor problem; learning; demand distribution; purchase decision 1. Introduction In newsvendor theory, optimal order and expected profit are functions of (1) price—the item’s purchase or selling price, and the salvage price; and (2) demand distribution (Nahmias, 1994, 2005). Previous literature tends to assume that the decision-maker faces know demand distribution or estimated distribution. Schweitzer and Cachon (2000) suggest that the assumption that the decision-maker knows the distribution of demand ‘is a reasonable assumption when the decision-maker has access to a substantial amount of historical data for similar products’ (p. 405). Fisher and Raman (1996) discovered that even though a fashion apparel manufacturer changes styles each year, the demand distribution for similar styles closely resembles that of previous years, although admittedly this observation refers to a very specific context of application. While it is possible that with enough historical data people may reasonably approximate the type of distribution and estimate its parameters, this is not always the case in real- life scenarios. There are many real-life instances where the demand distribution is unknown. This is especially true for the newsvendor problem, which is a single-period problem that includes one time occasions, specific holiday accessories, and special projects for which demand cannot be accurately projected. However, demand may be obscured even in other inventory models. For example, in presenting a new product to a new market, the vendor might face unknown demand distribution since s/he has no access to historical data for ∗ Correspondence: T Shavit, School of Business Administration, College of Management, 7 Rabin Avenue, Rishon-Le’Zion, Israel. E-mail: shavittal@gmail.com similar products. Even opening new business premises in a new area is a situation where the demand distribution is not known in advance. This paper tests experimentally, the order policy in the newsvendor problem in two cases: (1) where decision-makers know the demand distribution, and (2) where decision-makers do not know the demand distribution. Based on earlier liter- ature, it is assumed that decision-makers in the former cate- gory will order quantities that are closer to the optimal order, than those in the latter category. In the presented computerized learning experiment, each participant assumed the role of a newsvendor and had to decide how many papers to order each day over 100 days. The participants were paid at the end of the experiment according to their gains in a random period. The main finding is that knowing the demand distribution does not necessarily help in discovering the optimal solution or in improving the subject’s profits. In both cases, subjects are asymptotically converting to a personal order quantity that is different from the optimal one. Another important issue is the effect of feedback on the subjective order. The feedback in this case is the gap between the actual demand realization and the order of the previous day. It was found that this gap plays an important role in the participants order decisions. When the subject’s order is greater than the subsequent demand realization (a supply surplus) s/he has to throw the old newspapers away, absorbing a financial loss. Large losses are an incentive to be more cautious over the next period (ie reducing the order). On the other hand, when the subject orders less than the demand (a demand surplus) s/he sells all the ordered products and as a result faces only positive gain. While subjects may feel that