M. Sinan Gönül's research while affiliated with Northumbria University and other places

Publications (15)

Article
Full-text available
Forecasting has always been at the forefront of decision making and planning. The uncertainty that surrounds the future is both exciting and challenging, with individuals and organisations seeking to minimise risks and maximise utilities. The large number of forecasting applications calls for a diverse set of forecasting methods to tackle real-life...
Article
Full-text available
Forecasting has always been at the forefront of decision making and planning. The uncertainty that surrounds the future is both exciting and challenging, with individuals and organisations seeking to minimise risks and maximise utilities. The large number of forecasting applications calls for a diverse set of forecasting methods to tackle real-life...
Preprint
Full-text available
My contributions to this voluminous publication can be found on pp 38-40 "The natural law of growth in competition" and on pp 169-170 "Dealing with logistic forecasts in practice"
Chapter
The newsvendor problem is one of the rudimentary problems of inventory management with significant practical consequences, thus receiving considerable attention in the behavioral operational research literature. In this chapter, we focus on how decision makers perceive demand uncertainty in the newsvendor setting and discuss how such perception pat...
Article
Full-text available
Accurate forecasting is necessary to remain competitive in today's business environment. Forecast support systems are designed to aid forecasters in achieving high accuracy. However, studies have shown that people are distrustful of automated forecasters. This has recently been dubbed “algorithm aversion.” In this study, we explore the relationship...
Article
Full-text available
There is general agreement in many forecasting contexts that combining individual predictions leads to better final forecasts. However, the relative error reduction in a combined forecast depends upon the extent to which the component forecasts contain unique/independent information. Unfortunately, obtaining independent predictions is difficult in...
Article
This paper examines the accuracy of judgmental forecasts of product demand and the quality of subsequent production level decisions under two different conditions: (i) the availability of only time series information on past demand; (ii) the availability of time series information together with scenarios that outline possible prospects for the prod...
Article
In expert knowledge elicitation (EKE) for forecasting, the perceived credibility of an expert is likely to affect the weighting attached to their advice. Four experiments have investigated the extent to which the implicit weighting depends on the advisor's experienced (reflecting the accuracy of their past forecasts), or presumed (based on their st...
Article
Full-text available
The importance of advice taking in decision making has been arising for three decades. The existing research in advice taking has suggested numerous measurement techniques that are utilized under different types of advice such as random advice and perfect advice. In this study using the random advice framework, through an experimental design, we co...
Article
We present an experimental study of the price-setting newsvendor problem, which extends the traditional framework by allowing the decision maker to determine both the selling price and the order quantity of a given item. We compare behavior under this model with two benchmark conditions where subjects have a single decision to make (price or quanti...
Article
Today's business environment provides tougher competition than ever before, stressing the important role played by information and forecasts in decision-making. The scenario method has been popular for focused organizational learning, decision making and strategic thinking in business contexts, and yet, its use in communicating forecast information...
Article
Forecasts are important components of information systems. They provide a means for knowledge sharing and thus have significant decision-making impact. In many organizations, it is quite common for forecast users to receive predictions that have previously been adjusted by providers or other users of forecasts. Current work investigates some of the...
Article
Good decisions and policy depend on a reasonable degree of foresight; this in turn relies on a forecasting process that is well-integrated with the strategic management of an organization. There is, however, very little research into the forecast process within organizations generally, or health-services in particular. N.......Noting this gap, a fr...
Article
Research in the field of expert systems has shown that providing supporting explanations may influence effective use of system developed advice. However, despite many studies showing the less than optimal use made of DSS prepared advice, almost no research has been undertaken to study if the provision of explanations enhances the users' ability to...

Citations

... distributional forecasting. It did not go unnoticed to researchers [2,3], however the literature on probabilistic EPF is much scarcer than the one on point EPF. ...
... Thus, the assumption to supplement the database was based on the fact that the height of the trees cannot be known with certainty, even with conventional field measurements. So, the estimation by the previous model had a complementary role in the data's generation [32][33][34]. This allowed having the trajectories of the temporary plots to use the fit method, which requires more than one observation per plot [35,36]. ...
... Though it is widely established in the forecasting community that combining forecasts is beneficial, the gains from forecast combinations highly depend on several factors, including the quality of the pool of forecasts to be combined and the estimation of combination weights (Timmermann, 2006;Wang et al., 2022a). Naturally one would prefer to combine individual forecasts with high accuracy (Mannes et al., 2014;Kourentzes et al., 2019) and sufficient diversity (Batchelor & Dua, 1995;Thomson et al., 2019;Kang et al., 2022) to amplify the benefits of combinations. Alternatively, combination schemes have evolved from simple averaging without weight estimation to sophisticated methods tailoring weights for different individual forecasts. ...
... In practice, all too often garage managers make judgmental forecasts of the repair times based on a quick inspection. Such judgmental forecasts may not be consistent over time, and their accuracy highly depends on experience and skills ( Goodwin, Gonul & Onkal, 2019 ;Lawrence, Goodwin, O'Connor & Onkal, 2006 ). If repair times of automotive parts can be accurately predicted based on some formal approach, garage managers may considerably improve the efficiency of job scheduling and better inform customers about their waiting times. ...
... Indeed, this behaviorbelief (in)congruity is another dimension of algorithmic advice taking that is under-studied, and we believe that this may lead to false generalizations about the persuasive impact of algorithms. Past studies have found that statements or presumptions of trust in advisors do not necessarily correspond how influential their advice is in belief revision (Goodwin et al., 2013;Önkal et al., 2017). More specifically, higher levels of trust did not meaningfully increase persuasiveness compared to controls, though lower levels of trust did, albeit slightly. ...
... Prior studies on algorithm aversion hypothesized that there are a number of reasons for this preference for human-made forecast even in spite of explicit superiority of choices made by an algorithm: for instance, the notion that using algorithmbased choice modeling may be perceived as a loss of process ownership (Önkal and Gönül, 2005;Petropoulos et al., 2016), an abstract sense of unfamiliarity and hence distrust with the machine (Prahl and van Swol, 2017), or the notion that algorithms were unable to integrate qualitative factors (Grove and Meehl, 1996;Vrieze and Grove, 2009;Newman et al., 2020). Others argue that decision makers perceive algorithms as unable to account for unique and individual circumstances (Highhouse, 2008;Longoni et al., 2019), unable to deliver satisfying results in domains of high uncertainty (Dietvorst and Bharti, 2020), or mention machines' lack of intuition and fairness (Newman et al., 2020), a quality typically associated with human forecasting (Lee, 2018;Burton et al., 2020). ...
... Average orders placed are between the profit-maximizing order quantity and the mean demand, that is, orders are below (above) the optimum for products with a high (low) profit margin. This is called the pull-to-center effect (the center is the mean demand), and this order pattern has been found repeatedly in later studies (Bolton & Katok, 2008;Katok & Wu, 2009;Kocabiyikoglu et al., 2016;Moritz et al., 2013). Other deviations from optimality that have been found repeatedly are, among others, demand chasing and influence of targets (Benzion et al., 2008;Chen et al., 2015;Gavirneni & Xia, 2009;Minner & van Wassenhove, 2010). ...
... There are certain methods that have shown to produce higher quality scenario planning, as measured by clarity and confidence of practitioners, post hoc, and articulated action within affected communities (Cairns, Wright, & Fairbrother, 2016;Cairns, et al., 2017;Kuhn & Sniezek, 1996;Önkal, Sayim, & Gönül, 2013;Phadnis, et al., 2014;Schnaars & Topol, 1987). Suggestions are given for conducting scenario planning as interactive group sessions that consult a heterogeneous group of practitioners who bring a variety of knowledge and expertise that can be challenged. ...
... Automatic statistical baseline forecast: → Replacement with judgementally derived baseline forecast; → Further judgemental adjustment at Review Meeting to obtain final forecast (see Önkal, Gönül, and Lawrence (2008) for a laboratory study of how people adjust previously adjusted forecasts). ...
... The literature survey of related work on feature-based explanations reveals that they mainly vary in terms of three content design factors (Gönül et al., 2006;Chen and Pu, 2010;Chen and Chen, 2015;Sato et al., 2018): feature type, contextual relevance, and the number of features. Feature type refers to the type of feature to be explained, e.g., whether it primarily emphasizes the positive feature(s) of the recommended item (e.g., "delicious food" of a restaurant), or accommodates both positive and negative ones (called two-sided, e.g., "delicious food, but high price"). ...