No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Defining "Demand" for Demand Forecasting
Abstract
Demand forecasting is often uncritically based on histories of orders received, shipments/sales, or some combination of the two. As Mike Gilliland explains in this article, the ultimate goal - a measurement of true demand - is elusive and not always amenable to simple formulae based on orders and shipments. What to do? Recognizing the measurement difficulties, Mike suggests we can often derive a proxy for true demand that is close enough to be useful in generating an unconstrained forecast. Copyright International Institute of Forecasters, 2010
... Demand forecasting is one of the biggest challenges for retailers, wholesalers and manufacturers in any industry and it is a vital part of business intelligence (Kang et al., 2020;Khan et al., 2020;Pavlyshenko, 2019;Ren et al., 2019). Demand forecasting plays a very important role in marketing activities, especially sales & operations planning and supply chain management due to its demand volatility reducing effect (Ali et al., 2009;Bandara et al., 2019;Gilliland, 2010;Narayanan et al., 2019;Silveira-Netto and Brei, 2017;Fildes, et al., 2008;Khakpour, 2020). Accurate demand forecasting creates value in terms of predicting next purchases, avoiding under/over-stocking, sale opportunities, robust replenishment systems, new product planning, operational efficiency, customer satisfaction, competitive pricing and cost reductions (Chintagunta & Nair, 2011;Hu et al., 2019;Liu et al., 2013;Lu et al., 2012;Martinez et al., 2020;Trapero et al., 2015;Kumar et al., 2020;Yue et al., 2007). ...
In recent years, machine learning models based on big data have been introduced into marketing in order to transform customer data into meaningful insights and to make strategic decisions by making more accurate predictions. Although there is a large amount of literature on demand forecasting, there is a lack of research about how marketing strategies such as advertising and other promotional activities affect demand. Therefore, an accurate demand-forecasting model can make significant academic and practical contributions for business sustainability. The purpose of this article is to evaluate machine learning methods to provide accuracy in forecasting demand based on advertising expenses. The study focuses on a prediction mechanism based on several Machine Learning techniques—Support Vector Regression (SVR), Random Forest Regression (RFR) and Decision Tree Regressor (DTR) and deep learning techniques—Artificial Neural Network (ANN), Long Short Term Memory (LSTM),—to deal with demand forecasting based on advertising expenses. Deep learning is a powerful technique that can solve marketing problems based on both classification and regression algorithms. Accordingly, a television manufacturer’s real market dataset consisting of advertising expenditures, sales and demand forecasting via chosen machine learning methods was analyzed and compared in terms of the accuracy of demand forecasting. As a result, Long Short Term Memory has been found to be superior to other models in providing highly accurate prediction results for demand forecasting based on advertising expenses.
... Here are some prospects: Work to improve data quality and trustworthiness. For instance, refine measures of demands more precisely than is done through current data on shipments and orders (Gilliland, 2010). For example, when an item is out of stock the lost sales are not recorded or recorded at the wrong time. ...
The M5 forecasting competition is the latest and most widely contested since the first M competition in 1979. Numerous articles have been written appraising the structure of the competitions and the value of their results for forecasting methodology and practice. Mike Gilliland's discussion of the prior competition-the M4-in Foresight's Spring 2020 issue as well as Casper Bojer and Jens Peter Meldgaard's preview of the M5 in our Summer 2020 issue offer nontechnical overviews of these most recent forecasting competitions and their potential influence on the ways we forecast. Other background information on the competitions can be found at https://mofc.unic.ac.cy/. As there was with the M4, there will be a special issue of the International Journal of Forecasting-our sister IIF publication-devoted to a comprehensive assessment of the M5. The preprint "The M5 Accuracy Competition: Results, Findings, and Conclusions" provides a detailed discussion of the participants, methods, and results: https://www.researchgate. net/publication/344487258. Here, Spyros Makridakis, creator and overseer of the M forecasting competitions (and the person for whom they are named), and Evangelos Spiliotis, his closest collaborator, distill that initial report on the M5 to highlight the winning entries and what they say about the value of expertise in forecast modeling.
... Controlling the availability of services, i.e., efficiently managing an inventory to control "full price" sales is an important aspect (Curtis & Zahrn, 2014) and forecasting consists of estimating the KPIs of a hotel based on the defined segments for a given period, taking into account the micro and macroeconomic reality in which the hotel operates (Littlewood, 2005). Forecast helps hotels to organize themselves, since it allows managing the number of the needed employees for certain seasons or celebrations, investing in the necessary equipment, facilities and materials, depending on the demand forecast (Gilliland, 2010). Price is also one of the most important competencies of RM as it requires the ability to effectively evaluate products and services. ...
The world is constantly changing, and the hotel industry is one of the sectors where we can feel it the most. Nowadays, the hotel market is dynamic, diverse, difficult, and dangerous, which creates new challenges for hotel managers, particularly in terms of revenue, creation, and optimization. In order to overcome that, revenue management emerges as a crucial price management tool to face competitiveness and business growth. This study aims to analyze the practices and advantages of revenue management, as well as understand if its implementation in Portugal influenced the revenue and growth of the hotel industry. It also intends to analyze whether there is adequate revenue training or if further training should be required. A quantitative methodology was used, and 284 answers were collected. From those answers, 115 were validated, analyzed, and discussed. Conclusions were made, and finally some limitations were presented as well as suggestion for future research.
... Market sales and shipments from the manufacturer are the measured variables. Note that, unless company's fill rate is always 100%, either orders or sales do not reveal the true demand, which is unknown, and they are only estimations (Gilliland, 2010). Data from a manufacturing company specialized in household products have been collected. ...
Bullwhip effect is a problem of paramount importance that reduces competitiveness of supply chains around the world. A significant effort is being devoted by both practitioners and academics to understand its causes and to reduce its pernicious consequences. Nevertheless, limited research has been carried out to analyze potential metrics to measure it, that typically are summarized in the coefficient of variation ratio of different echelons demand.
This work proposes a new metric based on a time-varying extension of the aforementioned bullwhip effect metric by employing recursive estimation algorithms expressed in the State Space framework to provide at each single time period a real-time bullwhip effect estimate. In order to illustrate the results, a case study based on a serially-linked supply chain of two echelons from the chemical industry is analyzed. Particularly, this metric is employed to analyze the effect of promotional campaigns on the bullwhip effect on a real-time fashion. The results show that, effectively, the bullwhip effect is not constant along time, but interestingly, it is reduced during the promotional periods and it is bigger before and after the promotion takes place.
... Market sales and shipments from the manufacturer are the measured variables. Note that, unless company's fill rate is always 100%, either orders or sales do not reveal the true demand, which is unknown, and they are only estimations (Gilliland, 2010). Data from a manufacturing company specialized in household products have been collected. ...
Bullwhip effect is a problem of paramount importance that reduces com-petitiveness of supply chains around the world. A significant effort is being devoted by both practitioners and academics to understand its causes and to reduce its pernicious consequences. Nevertheless, limited research has been carried out to analyze potential metrics to measure it, that typically are summarized in the coefficient of variation ratio of different echelons demand. This work proposes a new metric based on a time-varying extension of the aforementioned bullwhip effect metric by employing recursive estimation al-gorithms expressed in the State Space framework to provide at each single time period a real-time bullwhip effect estimate. In order to illustrate the results, a case study based on a serially-linked supply chain of two echelons from the chemical industry is analyzed. Particularly, this metric is employed to analyze the effect of promotional campaigns on the bullwhip effect on a real-time fashion. The results show that, effectively, the bullwhip effect is not constant along time, but interestingly, it is reduced during the promotional periods and it is bigger before and after the promotion takes place.
... The supply chain system consists of a serially linked two-level supply chain, see Note that, unless company's fill rate is always 100%, either orders or sales do not reveal the true demand and they are only approximations [11]. Data from a manufacturing company specialized in household products have been collected. ...
The bullwhip effect (BE) consists of the demand variability amplification
that exists in a supply chain when moving upwards. This undesirable effect produces
excess inventory and poor customer service. Recently, several research papers from
either a theoretical or empirical point of view have indicated the nature of the demand
process as a key aspect to defining the BE. Nonetheless, they reached different
conclusions. On the one hand, theoretical research quantified the BE depending on
the lead time and ARIMA parameters, where ARIMA functions were employed to
model the demand generator process. In turn, empirical research related nonlinearly
the demand variability extent with the BE size. Although, it seems that both results
are contradictory, this paper explores how those conclusions complement each other.
Essentially, it is shown that the theoretical developments are precise to determine the
presence of the BE based on its ARIMA parameter estimates. Nonetheless, to quantify
the size of the BE, the demand coefficient of variation should be incorporated.
The analysis explores a two-staged serially linked supply chain, where weekly data
at SKU level from a manufacturer specialized in household products and a major
UK grocery retailer have been collected.
... In fact, there can be a substantial di erence between true demand and sales. As Mike Gilliland explained in his 2010 Foresight article (Gilliland, 2010), true demand may never be known exactly. ...
Misforecasting demand will cost profits and threaten reputations, market share, and even the business itself. Roland Martin and Stephan Kolassa draw on their expertise with retail companies to examine the major challenges for demand forecasting and the choices the demand forecaster must make. Their consideration of the options provides a valuable guidepost to the demand forecaster in measuring demand, handling questionable data, making adjustments to automatic forecasts, and choosing appropriate performance metrics.
In this chapter, the authors find contrasting views of the potential for big data in forecasting. They focus on recent statements made by software vendor Blue Ridge, statements that are quite typical in content for other market buzz on big data. Blue Ridge is suggesting that instead of forecasting by item, it is more sensible and ultimately more accurate to forecast total volume by customer and that big data is what makes this possible. Blue Ridge is making a very bold claim that big data is going to transform forecasting to be far more customer based. Blue Ridge is expressing a preference for using causal forecasting over time‐series statistical forecasting. Big data adds an additional capability – data mining – to identify causal factors. Firms are making significant investments in big‐data storage and applications. The traditional demand‐planning process begins with historical sales data and identifies potential explanatory variables.
ResearchGate has not been able to resolve any references for this publication.