An exponential smoothing model for predicting traffic in the library and at the reference desk

Abstract

PurposeThe purpose of this study is to develop a mathematical model that can be used to forecast the number of individuals who enter the library as well as the number of patrons that seek assistance at the reference desk of the library. An accurate estimate of demand at the reference desk is valuable for effective staffing decisions.Design/methodology/approachAn exponential smoothing model (Winter's model) was developed for forecasting. Data were gathered at the University of Tennessee at Chattanooga for an entire year. Using these data, an exponential smoothing model was formulated for forecasting the number of patrons seeking assistance. Since the data showed no trend but the presence of two seasonality factors, one for the week‐of‐the‐semester effect and one for the day‐of‐the‐week effect, the Winter's method appeared to be best suited. The Winter's method develops a formula from the data and allows the formula to be continuously fine‐tuned as new observations come in day after day.FindingsThe modified Winter's exponential smoothing model proved to be a good predictor of the number of patrons seeking assistance. In spite of large natural random variability present in the data, the actual values seem to follow the forecasts very closely.Originality/valueIt is vital to be able to forecast the number of clients at the reference desk that seek assistance per day. The modified exponential smoothing model is a valuable tool for such forecasting.

Academic library as learning space and as collection: A learning commons' effects on collections and related resources and services

“…Using survey data from 182 library building projects, Shill and Tonner concluded that "in general, building improvements had a greater overall impact on basic facility use (gate count) than on circulation, reference, transaction volume, and in-house collection use" (2004, p. 127). Other researchers have offered predictive models for calculating reference transactions based off gate counts and occupancy rates (Ahmadi, Dileepan, & Murgai, 2012;Ahmadi, Dileepan, Murgai, & Roth, 2008;Murgai & Ahmadi, 2007), but as none of the studies have considered the large spike in traffic associated with opening a learning commons, it is uncertain how applicable these models would be for libraries with these spaces.…”

Shall we chat? A statistical case study of chat reference utilization

“…A variety of quantitative and statistical methods have been applied to evaluate different aspects of reference services. Major topics considered include staffing needs (Murgai, 2006;Ryan, 2008;Applegate, 2008) and forecasting reference desk traffic flows (Murgai and Ahmadi, 2007;Ahmadi et al, 2008;Dubnjakovic, 2012). Statistical methods used in these studies include comparisons of means (Ryan, 2008), chi-squared residual analysis (Murgai, 2006;Applegate, 2008), multivariate regressions (Murgai and Ahmadi, 2007;Dubnjakovic, 2012) and time series analysis (Ahmadi et al, 2008).…”

“…Yet, to our knowledge, no quantitative study has been conducted so far to investigate the possible impact brought by a new reference tool, such as online chat, on the traditional face-to-face reference services. Ryan, 2008), chi-squared residual analysis (e.g., Murgai, 2006;Applegate, 2008), multivariate regressions (e.g., Murgai and Ahmadi, 2007;Dubnjakovic, 2012), and time series analysis (e.g., Ahmadi et al, 2008). Although in-depth qualitative assessment has been adopted in much of the existing research on chat reference, quantitative analysis of chat volume in academic libraries with statistical methods employed in other reference fields has rarely been conducted.…”

Predicting demand for inter-library loan requests

“…Mathematical forecasting models have been used for forecasting numerous variables in libraries. For instance, a multiple regression has been used to successfully predict the reference desk staffing requirements (Murgai and Ahmadi, 2007), and an exponential smoothing model was used to predict traffic in the library and at the reference desk (Ahmadi et al, 2008). In the current study, a Centered Moving Average with seasonal variation model was formulated for forecasting the demand for ILL.…”