Forecasting of Outbound Product Flow

Size: px
Start display at page:

Download "Forecasting of Outbound Product Flow"

Transcription

1 University of Twente Master s Thesis Extended Summary Forecasting of Outbound Product Flow Author: Catharina Elisabeth Lolkema Supervisory Committee: Dr. ir. L.L.M. van der Wegen Dr. M.C. van der Heijden J.J.T. Jaspers M.Sc. November 2, 2015

2 Author C.E. (Karin) Lolkema Education University of Twente M.Sc. Industrial Engineering and Management Specialization: Production and Logistic Management Graduation Date 6 November

3 Extended Summary This research study originates at the Supply Chain department of a global logistics service provider. Since 2005, logistical activities are employed for its only dedicated customer. One of these logistical processes, which we further refer to as the outbound process, is a highly intensive labor process directly subject to high fluctuating customer demand. Sharp standards are set by the customer to meet this fluctuating customer demand in time: received orders should be processed at the same day. To fulfill customer orders and manage peaks and falls in demand in the most effective manner against lowest possible cost, personnel should be scheduled efficiently one day in advance, what makes a reliable forecasting model as input for the personnel planning fundamental. In 2011, a forecasting model has been initiated for this outbound process by a graduation student, which currently retains a mean average absolute deviation of percent per day over the first 9 weeks of Besides its inaccuracy, forecasts generated by the model cannot be used as input for the personnel planning model due to the following reasons: the order structure in the forecasting model does not coincide with the order structure as required in the personnel scheduling model, and processes are differently measured in both models (items or lines), which cannot be directly rewritten into one another. Moreover, order types currently embedded in the model do not exist anymore, or new order types are not enclosed, leaving the model to retain a non up-to-date structure. Furthermore, nobody in the organization fully understands the model and has the ability to maintain it. Research Goal We identified the current forecasting model as quite inaccurate and costly, and specified the need for a forecasting model that can serve as an input for the personnel planning model. Since the planning of personnel occurs on a daily basis by the outbound (assistant) supervisor with an average educational background, our forecasting model should provide estimates on the same daily basis, and being easy to handle and maintain. Our research goal needs to satisfy these needs. In general terms, our main research goal is: Developing a practical forecasting model for the outbound process that is easy to understand and maintain, which serves as an input for the personnel planning model and provides forecasts on a daily level as accurate as possible. Note that our model aims to support the (assistant) supervisors in their scheduling proceedings, not to overrule their experiences. Since forecasting is only used for personnel scheduling purposes, the model 3

4 can be fully designed as input for the personnel planning model. To be clear, our new forecasting model will be newly developed, starting from scratch. Requirements Besides the persuasiveness to develop an easy-access forecasting model being an input for the personnel scheduling model, more requirements are defined. We adopt the outbound order structure from the personnel planning into our model, consisting of six outbound sub-processes with appropriate work load drive (WLD). The work load drive (lines or items) expresses the manner in which processes are measured. When work is embroiled per line, e.g. scanning of one box with 50 products (1 line with 50 items), the calculation of required employees is based on lines. When work is measured per item, the number of employees is calculated according to items. Forecasts should be finalized on a daily basis before the scheduling and recruitment process of personnel for the next day starts. As said earlier, received orders should be processed at the same day, meaning that today s volumes cannot be used for tomorrow s forecasts as they are not finalized yet at the time of scheduling. Accordingly, we propose to use yesterday s volumes for tomorrow s forecasts: a 2-day ahead forecast. To support this 2-day ahead forecast, we only opt for forecasting techniques useful for generating short-term forecasts. Ever changing order structures are faced by the customer due to marketing activities and changing customer preferences. As a result, the order structure should be adjustable in the sense that order types can easily be added and distracted, preferably by the people attaining such practical knowledge: the (assistant) supervisors. Last, to ease its maintenance and increase its sustainability within the organization, the forecasting model and inherent techniques should be easy to understand and maintain, requiring minimum data storage and running time. Selecting Forecasting Techniques To identify appropriate forecasting techniques, we initiated a literature research. Since our research is blessed with the availability of reliable historical data and we assume that patterns from the past will continue in the future, we narrow down our literature research to quantitative time series models. Type of Technique Technique Data pattern Naive Naive ST, T, S Averaging Moving averages ST Double moving averages ST, T Smoothing Simple exponential smoothing ST Trended exponential smoothing T Seasonal exponential smoothing T, S Table 1: A summary of the forecasting techniques and the data patterns they incorporate we use in our research, based on a framework provided by Hanke and Reitsch (1998), and Makridakis and Wheelwright (1978). Pattern of the data: ST, stationary; T, trended; S, seasonal; C, cyclical. 4

5 All quantitative time series models we identified during our literature research being appropriate for our situation are tabulated in Table 1. These models are ultimately suitable for short-term forecasts, easy to understand and maintain, easy to implement, and requiring minimum data storage and computational effort. With easy to understand and maintain we refer to forecasting models pertaining an easy initialization and update of parameters, that have a logical built-up of components to people without any statistical or mathematical background. According to Hanke and Reitsch (1998), the behavior of the data patterns mentioned in Table 1 can be defined as stated below. When making a forecast, the projection for each of these individual components is combined. Level (A): The level captures the scale of a time series, and shows a constant line if no trend is present. It indicates the initial height of a time series. Trend (T): The trend represents the growth or decline in the time series over a certain period of time. Seasonal component (S): The seasonal component is a pattern of change that repeats itself in a certain period of time (e.g. a week, month, or year). Techniques Although all techniques are relatively simple, some are simpler than others. We confine ourselves at this point to smoothing methods. Further details on the naive method and moving averages can be found in Hanke and Reitsch (1998) and Makridakis et al. (1978). Seasonal component Trend component N (none) A (additive) M (multiplicative) N (none) N-N N-A N-M A (additive) A-N A-A A-M DA (damped additive) DA-N DA-A DA-M M (multiplicative) M-N M-A M-M DM (damped multiplicative) DM-N DM-A DM-M Table 2: Exponential Smoothing Taxonomy Exponential Smoothing is very suitable in situations where managers need forecasts in a relatively short time. Taylor (2003) developed a taxonomy that describes various exponential smoothing techniques and their exploitation of different combinations of the trend and seasonal component, illustrated in Table 2. As can be seen, both the trend and seasonal component can be composed in an (damped) additive or (damped) multiplicative manner. For every combination of trend and seasonal component, an exponential smoothing method exists. Since an appropriate fit of a particular exponential smoothing technique with data gains surprisingly in accuracy, the detection of the best smoothing technique for our research is an important task. Based on the literature, we have chosen to incorporate the exponential smoothing methods listed in Table 3 in our research. These methods include Holt s, Brown s, and Winters models, which have been very popular with researchers and practitioners, and have proven their appropriateness and robustness in practice. Furthermore, we include the exponential trended models from Pegels (1969) and Taylor (2003), which are less commonly used, but have proven to be very accurate in a research study from Taylor as they tend to deal with the more common multiplicative data patterns (Hyndman et al., 2002; Taylor, 2003). 5

6 Trend Component N A DA M DM Seasonal Component N A M Simple Method At = α Yt + (1-α) Ŷt Ŷt+p = At Holt s Method At = αyt + (1-α)(At 1 + Tt 1) Tt = β(at - At 1) + (1-β)Tt 1 Ŷt+p = At + ptt Holt s Damped Method At = αyt + (1-α)(At 1 + ϕtt 1) Tt = β(at - At 1) + (1-β)ϕTt 1 Ŷt+p = At + p i=1 ϕ i Tt Pegels Method At = αyt + (1-α)(At 1Rt 1) Rt = β( A t At 1 + (1-β)Rt 1 Ŷt+p = AtR p t Pegels Damped Method At = αyt + (1-α)(At 1R ϕ t 1 ) Winters Additive Method At = α(yt St L) + (1-α)(At 1 + Tt 1) Tt = β(at - At 1) + (1-β)Tt 1 St = γ(yt At) + (1-γ)St L Ŷt+p = At + ptt + St L+p Winters Damped Additive Method At = αyt St L + (1-α)(At 1 + ϕtt 1) Tt = β(at - At 1) + (1-β)ϕTt 1 St = γ(yt At) + (1-γ)St L Ŷt+p = At + p i=1 ϕ i Tt + St L+p Winters Multiplicative Method At = α( Y t ) + (1-α)(At 1 + Tt 1) St L Tt = β(at - At 1) + (1-β)Tt 1 St = γ( Y t At ) + (1-γ)St L Ŷt+p = (At + ptt)st L+p Winters Damped Multiplicative Method At = α( Y t ) + (1-α)(At 1 + ϕtt 1) St L Tt = β(at - At 1) + (1-β)ϕTt 1 St = γ( Y t At Ŷt+p = (At + ) + (1-γ)St L p i=1 ϕ i Tt) St L+p Rt = β( A t A ϕ t 1 Ŷt+p = AtR + (1-β)Rt 1 p i=1 ϕi t Table 3: Exponential smoothing techniques and their equations used in our research. 6

7 Variable Y t Ŷ t+p α β γ ϕ A t T t R t S t L Description Observed value of the time series in period t Forecast for p periods ahead from period t Smoothing parameter for the level of a time series Smoothing parameter for the trend Smoothing parameter for the seasonal indices Autoregressive or damping parameter Smoothed level of a time series at the end of period t Smoothed additive trend of a time series at the end of period t Smoothed multiplicative trend of a time series at the end of period t Smoothed seasonal index of a time series at the end of period t Number of periods in the seasonal cycle Table 4: Explanation of variables used in the exponential smoothing taxonomy from Taylor (2003). Forecasting Forecasting involves the study of historical data to discover their underlying tendencies and patterns. This knowledge is used to project the data into future time periods as forecasts. (Hanke and Reitsch, 1998) To discover such tendencies and patterns, we first perform a data analysis. It helps us understanding if our chosen forecasting techniques form a good fit with our data and utilize existing data patterns to the fullest. Throughout this section, we use outbound process A as an illustrative example. Note that fictional data is used for process A: all facts and figures are made-up. Data analysis To involve timely data and follow Hyndman and Kostenko s (2007) suggestion that a minimum sample size of m+5 (where m stands for amount of seasons per year) observations is needed when data seems to own a yearly season, we opt for data from January 2012 till December 2014 for our data analysis. In respective order, per outbound sub-process we plot and analyze the data, perform autocorrelation analyses and use the Ljung-Box and t-statistic to strengthen our findings in estimating the existence of any (linear) trend or seasonality. Having put all these procedures into motion, we found that outbound process A possesses a negative linear trend and a weekly seasonality with a peaks in demand every Monday. Initial values To enable the forecasting process, initial values should be set for all parameters inherent in the forecasting techniques from Table 3: values for the data components A 0, T 0, S 0, and R 0, and values for the smoothing parameters α, β, γ and ϕ (only applicable for smoothing methods). We use the first 30 data points of 2012 to calculate the initial values for all data components, and use the (rest) data of 2012 and 2013 to set the initial values for the smoothing parameters, as our research has shown a higher forecast accuracy level can be achieved when more data is involved in setting these parameters. For both types of parameters (data component, smoothing parameter), different calculations are used. Data components are valued by the pragmatic and popular ratio to moving average technique. 7

8 Smoothing parameters are set by use of a grid search in Microsoft Excel s SOLVER. The SOLVER function minimizes the quadratic differences between observed values and forecasted values (error) in terms of the mean squared error (MSE). A minimization of the MSE measure is especially suitable for our setting, as it induces large penalties to large forecast errors by squaring all errors, which is useful in need of an accurate personnel planning every single day. We only allow values for the smoothing parameters between 0 and 1, as this range is often used in practice and more restricted parameters do not necessarily improve forecast accuracy (Gardner, 1985). Microsoft Excel s SOLVER is chosen because of its fast working and level of detail. Aggregation level, time horizon and outliers This section details some further requirements to include in the forecasting process. At first, we remark that forecasting can take place at two levels: (1) per outbound order type separately, with an addup of forecasts afterwards, or (2) directly at an aggregated level with an add-up of volumes per order type prior to forecasting. We decided to pursue the latter option, because analysis pointed out that on average a 1.12 percent gain in forecast accuracy can be reached. Furthermore, to make inclusion of marketing information and verification of volumes easier by our logistics service provider s customer, a weekly verification would be more appealing than a daily verification. Research showed that an increase of 4.11 percent in forecast accuracy can be reached when organizing a daily verification. As our research aims to achieve the highest possible accuracy level (and a daily verification is a reasonable prospect), we produce a 2-day ahead forecast on the aggregated level of an outbound sub-process. Last, as we do not want our forecasting techniques to keep track of non-repetitive events far into the future, we remove these events in the form of outliers from our data set. As its parameters used for outlier calculation are not affected by outliers, we use the Modified Z-Score technique for outlier detection (Moore and McCabe, 2002). Moreover, the technique is especially useful for our situation as it does not ask for any kind of prior knowledge or assumptions on outliers from the (assistant) supervisors, and does not acquire large storage capacity or incremental running time. Best forecasting technique Having set initial values and having applied the requirements from the previous section in our research, we can take the next step: detecting the best forecasting technique per outbound sub-process based on a proper fit with our data and delivered forecast accuracy. A decent fit of a forecasting technique with data is important to actually forecast and utilize existing patterns to the fullest. We identify a proper fit by plotting both the observed values and forecasts over 2014, and using an Autocorrelation Function (ACF) to discover any remaining pattern in the residuals (difference between the observed value and forecast). Furthermore, we use the property of smoothing parameters that they form a good fit if they are valued between 0.1 and 0.3 (Jacob and Chase, 2013; Schroeder et al., 2013; Gardner, 1985). Figure 1 visualizes the plots of both the observed values and forecasts, in which the simple exponential smoothing and Winters multiplicative exponential smoothing are exemplified. As can be seen, forecasts generated by Winters technique with additional seasonal parameter form a better fit with the data than the exponential smoothing technique with no trended nor seasonal parameter. 8

9 Figure 1: Behavior of forecasts generated by the simple smoothing and Winters multiplicative smoothing techniques in comparison to the observed values from Process A in If the Autocorrelation Function (ACF) of residuals appear not be random, a certain pattern still exists in the residuals, meaning that the forecasting technique does not properly utilize present data patterns (Hanke and Reitsch, 1998). Figure 2 proposes the ACF s of the residuals generated by all forecasting techniques. Only the autocorrelations for the naive method and damped Winters multiplicative appear to be randomly distributed, meaning they form a proper fit with the observed values as no pattern is left in the residuals. 9

10 Figure 2: Autocorrelation Functions (ACF) of the residuals produced by all forecasting techniques during the forecasting process for outbound Process A. Confining ourselves to the last step in identifying a proper fit, Table 5 tabulates the range of values the smoothing parameters have taken on during the forecast process. As smoothing values above 0.3 indicate an overcompensation of the parameter to keep up with patterns they are not built for, we conclude damped Holt s and damped Winters additive smoothing techniques to not form a proper fit with our observed values. Furthermore, techniques with smoothing values valued below 0.1 neither simulate a perfect fit, meaning that on these grounds, Brown s and Holt s are discarded too. 10

11 Parameters Forecasting Technique α β γ ϕ Simple Exponential Smoothing 0.25 Brown s Exponential Smoothing 0.03 Holt s Exponential Smoothing Holt s Damped Exponential Smoothing Winter s Mult. Exponential Smoothing Winter s Mult. Damped Exponential Smoothing Winter s Add. Exponential Smoothing Winter s Add. Damped Exponential Smoothing Pegel s Exponential Smoothing Pegel s Damped Exponential Smoothing NA NA NA NA Table 5: Parameter setting for the exponential smoothing techniques; used to generate forecasts for Process A. Excluding the forecasting techniques per outbound sub-process that do not form a good fit with its observed values, we submit remaining forecasting techniques to forecast accuracy measurements in terms of mean squared error (MSE), mean absolute deviation (MAD), and mean average percentage error (MAPE). From all techniques forming a good fit, the damped Winters multiplicative exponential smoothing technique appears to generate most accurate forecasts for outbound Process A. Table 6 tabulates the most appropriate forecasting technique per outbound sub-process, retrieving highest accuracy levels in all terms. We have opt for one generic forecasting technique for all outbound processes, but forecast accuracy would then be decreased by at least 8.7 percent (when generalizing Winters multiplicative method) on average for all processes, which is not consistent with our research goal to strive for the best possible result. Outbound sub-process Process A Process B Process C Process D Process E Process F Forecasting technique Damped Winters multiplicative exponential smoothing Damped Winters additive exponential Smoothing Damped Holt s exponential smoothing Winters additive exponential smoothing Winters multiplicative exponential smoothing Damped Winters additive exponential smoothing Table 6: An overview of the most appropriate forecasting technique per outbound sub-process. Further improvements To further improve forecast accuracy, we identified the influence of national holidays on its observed values just before or just after, and investigated the appropriate time at which to update the smoothing parameters (α, β, γ, ϕ) in smoothing methods. Adding prior knowledge on influences of national holidays in our data samples, we improved the forecast accuracy for Process A with 0.02 percent on average per day. By use of a tracking signal, which signals a drift in forecasts by means of a confidence interval, with an update at a 99 percent service level, we improved the forecast accuracy by 0.24 percent on average per day, in relation to no update at all. 11

12 Implementation Our forecasting model is built in Microsoft Excel, since this software programme is available on all computers within the organization, and general knowledge of the software is available in-house. Additionally, investment costs are saved. Our research has partially been initiated because of a lack of proper understanding of the current forecasting model and a non-defined responsibility. We therefore need to be aware of common failures of forecasting models, like a low credibility (Makridakis and Wheelwright, 1979), and low organizational support (Sanders and Manrodt, 1994). According to Silver et al. (1998), to improve credibility, one must be able to understand underlying choices and assumptions in the forecasting model, and the general idea of the statistical model. We therefore find it important to transfer knowledge about the model properly to the (assistant) supervisors, and make them responsible for the generation of accurate forecasts, merely because they attain most up-to-date information on operational proceedings. For additional proceedings, like the add-in of new processes, we suggest the Supply Chain department provides support, as they have knowledge of Microsoft Excel and logistical processes. To support a proper implementation, we recommend to train all involved people that need to work with the model: the (assistant) supervisors and the Supply Chain department. Clear manuals should be written, most preferable with an outline of the actions to take in case an error or other kind of scenario occurs. Besides a proper transfer of knowledge, any risk in implementation is located in IT capacity. Results Carrying out all our findings, we retain an average percentage deviation of actual volumes from forecasted volumes of percent per outbound sub-process per day. In contrast, the old forecasting model released an average percentage deviation of percent, which means we achieved an improvement of percent point on average per day. Conclusions and Recommendations We draw the following conclusions: We conclude that our research has delivered a forecasting model which satisfies our research goal. It not only improves the forecast accuracy by percent point on average per day, the model is also easy to understand and maintain by the (assistant) supervisor(s), as order types and its belonging historical data can easily be added and distracted from the total demand streams per process. Furthermore, the forecasting techniques used in the model can easily be interpreted by employees of the Supply Chain department to provide additional support (like the add-in of new processes). Besides, all techniques take on an easy update of parameters, without incremental running time. To increase its sustainability within the organization, the model is built in Microsoft Excel, which is accessible on all computers and of which general knowledge is available in-house. Furthermore, investment costs are saved. Most importantly, the model is customized towards the personnel planning, as it retains the same order structure and provides forecasts on a daily basis, to support the supervisors in their forecasting and planning process. 12

13 Of course, there is always room for improvement: Relating to the positive results, we recommend to use the new forecasting model as an input for the personnel planning. To finalize the bridge towards practice, the supervisors should own the model, as they now have the ability to properly work with the model. For additional proceedings, like the add-in of new order types, the Supply Chain department should be able to lend a hand. To sustain optimal results in case new order outbound processes are added, we propose further research on an automatic identification of the most appropriate forecasting technique per process. To include marketing information in forecasts and achieve even more accurate results, a verification process by the service provider s customer, who s attaining such knowledge in-house, should be introduced. Bibliography Gardner, E.S. (1985). Exponential Smoothing: The State of the Art. Journal of Forecasting, Vol. 4, Hanke, J.E., and Reitsch, A.G. (1998). Business Forecasting. 6th edition, Prentice Hall. Hyndman, R.J., Koehler, A.B., Snyder, R.D., Grose, S. (2002). A state space framework for automatic forecasting using exponential smoothing methods. International Journal of Forecasting 18, Hyndman, R.J., and Kostenko, A.V. (2007). Minimum sample size requirements for seasonal forecasting models. Foresight, Issue 6. Jacobs, F.R., and Chase, R.B. (2013). Operations and Supply Chain Management: The Core. 3rd Edition, McGraw-Hill Higher Education, Page 59, Chapter 3. Makridakis, S., and Wheelwright, S.C. (1978). Forecasting, methods and applications. 1st edition. USA: John Wiley and Sons Inc. Makridakis, S., and Wheelwright, S.C. (1979). Forecasting the Future and the Future of Forecasting. In S. Makridakis and S.C. Wheelwright (eds.), TIMS Studies in the Management Sciences, vol 12. Amsterdam: North-Holland, pp Moore, D.S., and McCabe, G.P. (2002). Introduction to the Practice of Statistics. 4th ed. New York: W. H. Freeman,. Pegels, C. (1969). Exponential forecasting: some new variations. Management Science, 15, Sanders, N., and K. Manrodt (1994). Forecasting Practices in US Corporations: Survey Results. Interfaces, 24(2), Schroeder, R., Rungtusanatham, M. J., and Goldstein, S. (2013). Operations Management in the Supply Chain: Decisions and Cases. 6th Edition, McGraw-Hill Higher Education, Page 261, Chapter 11. Silver, E.A., Pyke, D.F., Peterson, R., (1998), Inventory Management and Production Planning and 13

14 Scheduling. 3th edition, USA: John Wiley and Sons Inc. Taylor, J.W. (2003). Exponential smoothing with a damped multiplicative trend. International Journal of Forecasting, 19 (2003)

COORDINATING DEMAND FORECASTING AND OPERATIONAL DECISION-MAKING WITH ASYMMETRIC COSTS: THE TREND CASE

COORDINATING DEMAND FORECASTING AND OPERATIONAL DECISION-MAKING WITH ASYMMETRIC COSTS: THE TREND CASE COORDINATING DEMAND FORECASTING AND OPERATIONAL DECISION-MAKING WITH ASYMMETRIC COSTS: THE TREND CASE ABSTRACT Robert M. Saltzman, San Francisco State University This article presents two methods for coordinating

More information

SIOPRED performance in a Forecasting Blind Competition

SIOPRED performance in a Forecasting Blind Competition SIOPRED performance in a Forecasting Blind Competition José D. Bermúdez, José V. Segura and Enriqueta Vercher Abstract In this paper we present the results obtained by applying our automatic forecasting

More information

Enhancing Forecasting Capability of Excel with User Defined Functions

Enhancing Forecasting Capability of Excel with User Defined Functions Spreadsheets in Education (ejsie) Volume 2 Issue 3 Article 6 5-10-2008 Enhancing Forecasting Capability of Excel with User Defined Functions Deepak K. Subedi Marshall University, subedi@marshall.edu Follow

More information

Managers require good forecasts of future events. Business Analysts may choose from a wide range of forecasting techniques to support decision making.

Managers require good forecasts of future events. Business Analysts may choose from a wide range of forecasting techniques to support decision making. Managers require good forecasts of future events. Business Analysts may choose from a wide range of forecasting techniques to support decision making. Three major categories of forecasting approaches:

More information

APPLICATION OF TIME-SERIES DEMAND FORECASTING MODELS WITH SEASONALITY AND TREND COMPONENTS FOR INDUSTRIAL PRODUCTS

APPLICATION OF TIME-SERIES DEMAND FORECASTING MODELS WITH SEASONALITY AND TREND COMPONENTS FOR INDUSTRIAL PRODUCTS International Journal of Mechanical Engineering and Technology (IJMET) Volume 8, Issue 7, July 2017, pp. 1599 1606, Article ID: IJMET_08_07_176 Available online at http://www.iaeme.com/ijmet/issues.asp?jtype=ijmet&vtype=8&itype=7

More information

AWERProcedia Information Technology & Computer Science

AWERProcedia Information Technology & Computer Science AWERProcedia Information Technology & Computer Science Vol 03 (2013) 370-374 3 rd World Conference on Information Technology (WCIT-2012) Automatically Forecasting Magazine Demands and Modeling Seasonality

More information

DEPARTMENT OF QUANTITATIVE METHODS & INFORMATION SYSTEMS

DEPARTMENT OF QUANTITATIVE METHODS & INFORMATION SYSTEMS DEPARTMENT OF QUANTITATIVE METHODS & INFORMATION SYSTEMS Time Series and Their Components QMIS 320 Chapter 5 Fall 2010 Dr. Mohammad Zainal 2 Time series are often recorded at fixed time intervals. For

More information

IBM SPSS Forecasting 19

IBM SPSS Forecasting 19 IBM SPSS Forecasting 19 Note: Before using this information and the product it supports, read the general information under Notices on p. 108. This document contains proprietary information of SPSS Inc,

More information

In Chapter 3, we discussed the two broad classes of quantitative. Quantitative Forecasting Methods Using Time Series Data CHAPTER 5

In Chapter 3, we discussed the two broad classes of quantitative. Quantitative Forecasting Methods Using Time Series Data CHAPTER 5 CHAPTER 5 Quantitative Forecasting Methods Using Time Series Data In Chapter 3, we discussed the two broad classes of quantitative methods, time series methods and causal methods. Time series methods are

More information

Demand Forecasting of Power Thresher in a Selected Company

Demand Forecasting of Power Thresher in a Selected Company International Conference on Mechanical, Industrial and Materials Engineering 215 (ICMIME215) 11-13 December, 215, RUET, Rajshahi, Bangladesh. Paper ID: IE-1 Demand Forecasting of Power Thresher in a Selected

More information

INTRODUCTION BACKGROUND. Paper

INTRODUCTION BACKGROUND. Paper Paper 354-2008 Small Improvements Causing Substantial Savings - Forecasting Intermittent Demand Data Using SAS Forecast Server Michael Leonard, Bruce Elsheimer, Meredith John, Udo Sglavo SAS Institute

More information

Determination of Optimum Smoothing Constant of Single Exponential Smoothing Method: A Case Study

Determination of Optimum Smoothing Constant of Single Exponential Smoothing Method: A Case Study Int. J. Res. Ind. Eng. Vol. 6, No. 3 (2017) 184 192 International Journal of Research in Industrial Engineering www.riejournal.com Determination of Optimum Smoothing Constant of Single Exponential Smoothing

More information

DEMAND FORECASTING FOR FERTILIZERS A TACTICAL PLANNING FRAMEWORK FOR INDUSTRIAL USE

DEMAND FORECASTING FOR FERTILIZERS A TACTICAL PLANNING FRAMEWORK FOR INDUSTRIAL USE Demand Forecasting for Fertilizers A Tactical Planning Framework for Industrial Use Proceedings of AIPA 2012, INDIA 123 DEMAND FORECASTING FOR FERTILIZERS A TACTICAL PLANNING FRAMEWORK FOR INDUSTRIAL USE

More information

AIR FORCE INSTITUTE OF TECHNOLOGY

AIR FORCE INSTITUTE OF TECHNOLOGY PHARMACEUTICAL INVENTORY FORECASTING AT THE WRIGHT-PATTERSON MEDICAL CENTER Patrick J. Reymann, Capt, USAF AFIT/GTM/LAL/95S-12 DTWQ UALnTmapBcigDa DEPARTMENT OF THE AIR FORCE AIR UNIVERSITY AIR FORCE INSTITUTE

More information

Selection of a Forecasting Technique for Beverage Production: A Case Study

Selection of a Forecasting Technique for Beverage Production: A Case Study World Journal of Social Sciences Vol. 6. No. 3. September 2016. Pp. 148 159 Selection of a Forecasting Technique for Beverage Production: A Case Study Sonia Akhter**, Md. Asifur Rahman*, Md. Rayhan Parvez

More information

Choosing Smoothing Parameters For Exponential Smoothing: Minimizing Sums Of Squared Versus Sums Of Absolute Errors

Choosing Smoothing Parameters For Exponential Smoothing: Minimizing Sums Of Squared Versus Sums Of Absolute Errors Journal of Modern Applied Statistical Methods Volume 5 Issue 1 Article 11 5-1-2006 Choosing Smoothing Parameters For Exponential Smoothing: Minimizing Sums Of Squared Versus Sums Of Absolute Errors Terry

More information

Forecasting Software

Forecasting Software Appendix B Forecasting Software B1 APPENDIX B Forecasting Software Good forecasting software is essential for both forecasting practitioners and students. The history of forecasting is to a certain extent

More information

(c) 2010, KnowledgeBased Systems Corporation

(c) 2010, KnowledgeBased Systems Corporation Mostly this presentation extends a previous work by Drs. Armsrong, Collopy and Adya, Published in 1992 but includes input from many of the references, books and white papers. Remember: Forecasting can

More information

Optimizing Production Process through Production Planning and Inventory Management in Motorcycle Chains Manufacturer

Optimizing Production Process through Production Planning and Inventory Management in Motorcycle Chains Manufacturer ComTech: Computer, Mathematics and Engineering Applications, 9(2), December 2018, 43-50 DOI: 10.21512/comtech.v9i2.4723 P-ISSN: 2087-1244 E-ISSN: 2476-907X Optimizing Production Process through Production

More information

Minimizing the Impact of Forecast Error on Government Monetary and Fiscal Policy through Forecasting Software

Minimizing the Impact of Forecast Error on Government Monetary and Fiscal Policy through Forecasting Software Volume Issue 3, 159-17, 17, ISSN:-319 85 Minimizing the Impact of Forecast Error on Government Monetary and Fiscal Policy through Forecasting Software Onu Fergus U. (Ph.D) Nnanna Emmanuel E. Department

More information

LECTURE 8: MANAGING DEMAND

LECTURE 8: MANAGING DEMAND LECTURE 8: MANAGING DEMAND AND SUPPLY IN A SUPPLY CHAIN INSE 6300: Quality Assurance in Supply Chain Management 1 RESPONDING TO PREDICTABLE VARIABILITY 1. Managing Supply Process of managing production

More information

Spreadsheets in Education (ejsie)

Spreadsheets in Education (ejsie) Spreadsheets in Education (ejsie) Volume 2, Issue 2 2005 Article 5 Forecasting with Excel: Suggestions for Managers Scott Nadler John F. Kros East Carolina University, nadlers@mail.ecu.edu East Carolina

More information

Operations Management

Operations Management 12-1 Aggregate Planning Operations Management William J. Stevenson 8 th edition 12-2 Aggregate Planning CHAPTER 12 Aggregate Planning McGraw-Hill/Irwin Operations Management, Eighth Edition, by William

More information

COORDINATING DEMAND FORECASTING AND OPERATIONAL DECISION-MAKING: RESULTS FROM A MONTE CARLO STUDY AND A CALL CENTER

COORDINATING DEMAND FORECASTING AND OPERATIONAL DECISION-MAKING: RESULTS FROM A MONTE CARLO STUDY AND A CALL CENTER COORDINATING DEMAND FORECASTING AND OPERATIONAL DECISION-MAKING: RESULTS FROM A MONTE CARLO STUDY AND A CALL CENTER Rech, Paul & Saltzman, Robert Information Systems & Business Analysis Department San

More information

A Parametric Bootstrapping Approach to Forecast Intermittent Demand

A Parametric Bootstrapping Approach to Forecast Intermittent Demand Proceedings of the 2008 Industrial Engineering Research Conference J. Fowler and S. Mason, eds. A Parametric Bootstrapping Approach to Forecast Intermittent Demand Vijith Varghese, Manuel Rossetti Department

More information

FORECASTING AND DEMAND MANAGEMENT

FORECASTING AND DEMAND MANAGEMENT FORBUS NOVEMBER 2013 EXAMINATION DATE: 8 NOVEMBER 2013 TIME: 09H00 11H00 TOTAL: 100 MARKS DURATION: 2 HOURS PASS MARK: 40% (XN-88) FORECASTING AND DEMAND MANAGEMENT THIS EXAMINATION PAPER CONSISTS OF 3

More information

DRAFT FOR DISCUSSION AND REVIEW, NOT TO BE CITED OR QUOTED

DRAFT FOR DISCUSSION AND REVIEW, NOT TO BE CITED OR QUOTED DRAFT FOR DISCUSSION AND REVIEW, NOT TO BE CITED OR QUOTED USING MISSPECIFICATION TESTS ON WITHIN-SAMPLE MODELS AND CONSIDERING THE TRADE-OFF BETWEEN LEVEL AND SIGNIFICANCE AND POWER TO PROMOTE WELL-CALIBRATED,

More information

Traditional Forecasting Applied to Retail Demand

Traditional Forecasting Applied to Retail Demand Traditional Forecasting Applied to Retail Demand Research Paper Business Analytics Cor Woudt Supervisor: Ger Koole November 2018 1 Abstract This paper elaborates on conventional traditional forecasting

More information

Expert Opinions About Extrapolation and the Mystery of the Overlooked Discontinuities

Expert Opinions About Extrapolation and the Mystery of the Overlooked Discontinuities University of Pennsylvania ScholarlyCommons Marketing Papers Wharton School December 1992 Expert Opinions About Extrapolation and the Mystery of the Overlooked Discontinuities Fred Collopy Case Western

More information

Big Data: Baseline Forecasting With Exponential Smoothing Models

Big Data: Baseline Forecasting With Exponential Smoothing Models 8 Big Data: Baseline Forecasting With Exponential Smoothing Models PREDICTION IS VERY DIFFICULT, ESPECIALLY IF IT IS ABOUT THE FUTURE NIELS BOHR (1885-1962), Nobel Laureate Physicist Exponential smoothing

More information

Optimizing Production Process through Production Planning and Inventory Management in Motorcycle Chains Manufacturer

Optimizing Production Process through Production Planning and Inventory Management in Motorcycle Chains Manufacturer ComTech: Computer, Mathematics and Engineering Applications, 9(2), December 2018, 43-50 DOI: 10.21512/comtech.v9i2.4723 P-ISSN: 2087-1244 E-ISSN: 2476-907X Optimizing Production Process through Production

More information

Operational Logistics Management (OLM612S)

Operational Logistics Management (OLM612S) Ester Kalipi (M.LSCM.; B. Hons Logistics; B-tech. BA.; Dip. BA.; Cert. BA.) Operational Logistics Management (OLM612S) Unit 2: Logistics Planning 21 February 2018 Table of contents Unit Objectives The

More information

New Methods and Data that Improves Contact Center Forecasting. Ric Kosiba and Bayu Wicaksono

New Methods and Data that Improves Contact Center Forecasting. Ric Kosiba and Bayu Wicaksono New Methods and Data that Improves Contact Center Forecasting Ric Kosiba and Bayu Wicaksono What we are discussing today Purpose of forecasting (and which important metrics) Humans versus (Learning) Machines

More information

Comparison of Efficient Seasonal Indexes

Comparison of Efficient Seasonal Indexes JOURNAL OF APPLIED MATHEMATICS AND DECISION SCIENCES, 8(2), 87 105 Copyright c 2004, Lawrence Erlbaum Associates, Inc. Comparison of Efficient Seasonal Indexes PETER T. ITTIG Management Science and Information

More information

A Comparative Study of Different Statistical Techniques Applied to Predict Share Value of State Bank of India (SBI)

A Comparative Study of Different Statistical Techniques Applied to Predict Share Value of State Bank of India (SBI) A Comparative Study of Different Statistical Techniques Applied to Predict Share Value of State Bank of India (SBI) Hota H.S., Sahu Pushpanjali Abstract. Prediction of share value is one of the critical

More information

Forecasting Process for Predicting Container Volumes in the Shipping Industry

Forecasting Process for Predicting Container Volumes in the Shipping Industry Master Degree Project in Logistics and Transport Management Forecasting Process for Predicting Container Volumes in the Shipping Industry Solmaz Darabi and Mirza Suljevic Supervisor: Rick Middel Master

More information

Synagogue Capacity Planning

Synagogue Capacity Planning Synagogue Capacity Planning Ellie Schachter Professor Lefkovitz JWSS Module 22 March 2017 Introduction 3 Current State 3 Data Analysis 4 Predictive Model 7 Future State 15 Works Cited 16 2 Introduction

More information

THE IMPROVEMENTS TO PRESENT LOAD CURVE AND NETWORK CALCULATION

THE IMPROVEMENTS TO PRESENT LOAD CURVE AND NETWORK CALCULATION 1 THE IMPROVEMENTS TO PRESENT LOAD CURVE AND NETWORK CALCULATION Contents 1 Introduction... 2 2 Temperature effects on electricity consumption... 2 2.1 Data... 2 2.2 Preliminary estimation for delay of

More information

IMPLEMENTING ECONOMIC ORDER INTERVAL FOR MULTI ITEM TO REDUCE TOTAL INVENTORY COST

IMPLEMENTING ECONOMIC ORDER INTERVAL FOR MULTI ITEM TO REDUCE TOTAL INVENTORY COST IMPLEMENTING ECONOMIC ORDER INTERVAL FOR MULTI ITEM TO REDUCE TOTAL INVENTORY COST Anastasia L. Maukar Faculty of Technology, Industrial Engineering Department, President University Jl. Ki HajarDewantara

More information

EFFICACY OF ROBUST REGRESSION APPLIED TO FRACTIONAL FACTORIAL TREATMENT STRUCTURES MICHAEL MCCANTS

EFFICACY OF ROBUST REGRESSION APPLIED TO FRACTIONAL FACTORIAL TREATMENT STRUCTURES MICHAEL MCCANTS EFFICACY OF ROBUST REGRESSION APPLIED TO FRACTIONAL FACTORIAL TREATMENT STRUCTURES by MICHAEL MCCANTS B.A., WINONA STATE UNIVERSITY, 2007 B.S., WINONA STATE UNIVERSITY, 2008 A THESIS submitted in partial

More information

SAP APO DP (Demand Planning) Sample training content and overview.. All rights reserved Copyright 2005 TeknOkret Services. All Rights Reserved.

SAP APO DP (Demand Planning) Sample training content and overview.. All rights reserved Copyright 2005 TeknOkret Services. All Rights Reserved. SAP APO DP (Demand Planning) Sample training content and overview Sample course content Demand Planning Concepts Importance of Demand Planning SAP APO Demand Planning terminology APO Data structure: Source

More information

Managing Stock Availability in East Malaysia

Managing Stock Availability in East Malaysia Managing Stock Availability in East Malaysia By Tabiash Shandab Thesis Advisor: Dr. Albert Tan Summary: The project analyzed current stock availability of the selected base products for East Malaysia and

More information

Integration of Demand Management in Production Planning and Purchasing Management: Metal Packaging Industry The Colep Case Study

Integration of Demand Management in Production Planning and Purchasing Management: Metal Packaging Industry The Colep Case Study Integration of Demand Management in Production Planning and Purchasing Management: Metal Packaging Industry The Colep Case Study Diogo Lopo Department of Engineering and Management, Instituto Superior

More information

Application of Theory of Constraint Supply Chain Replenishment System in Fast Moving Consumer Goods Company

Application of Theory of Constraint Supply Chain Replenishment System in Fast Moving Consumer Goods Company Vol. 6 No. 4, December 2017 167 Application of Theory of Constraint Supply Chain Replenishment System in Fast Moving Consumer Goods Company Margaretha 1, Dyah Budiastuti 2, Taufik Roni Sahroni 3 1,3 Master

More information

REGIONAL DEMAND FORECASTING STUDY FOR TRANSPORTATION FUELS IN TURKEY

REGIONAL DEMAND FORECASTING STUDY FOR TRANSPORTATION FUELS IN TURKEY REGIONAL DEMAND FORECASTING STUDY FOR TRANSPORTATION FUELS IN TURKEY Özlem Atalay Prof. Gürkan Kumbaroğlu INTRODUCTION The prediction of fuel consumption has been an important tool for energy planning,

More information

Analysis of Various Forecasting Approaches for Linear Supply Chains based on Different Demand Data Transformations

Analysis of Various Forecasting Approaches for Linear Supply Chains based on Different Demand Data Transformations Institute of Information Systems University of Bern Working Paper No 196 source: https://doi.org/10.7892/boris.58047 downloaded: 13.3.2017 Analysis of Various Forecasting Approaches for Linear Supply Chains

More information

MIT SCALE RESEARCH REPORT

MIT SCALE RESEARCH REPORT MIT SCALE RESEARCH REPORT The MIT Global Supply Chain and Logistics Excellence (SCALE) Network is an international alliance of leading-edge research and education centers, dedicated to the development

More information

Syllabus Autumn 2015 OPMGT 443: Inventory and Supply Chain Management M W 10:30-12:20 DEM 004

Syllabus Autumn 2015 OPMGT 443: Inventory and Supply Chain Management M W 10:30-12:20 DEM 004 Syllabus Autumn 2015 OPMGT 443: Inventory and Supply Chain Management M W 10:30-12:20 DEM 004 Prof. Apurva Jain apurva@uw.edu PACCAR 532, Ph: 685-4970; Office hours: M W 9:30-10:25 Course Objectives &

More information

Determining the Effectiveness of Specialized Bank Tellers

Determining the Effectiveness of Specialized Bank Tellers Proceedings of the 2009 Industrial Engineering Research Conference I. Dhillon, D. Hamilton, and B. Rumao, eds. Determining the Effectiveness of Specialized Bank Tellers Inder S. Dhillon, David C. Hamilton,

More information

Week 1 Business Forecasting

Week 1 Business Forecasting Week 1 Business Forecasting Forecasting is an attempt to foresee the future by examining the past, present and trends Forecasting involves the prediction of future events or future outcomes of key variables.

More information

Forecasting product demand for a retail chain to reduce cost of understocking and overstocking Group B5

Forecasting product demand for a retail chain to reduce cost of understocking and overstocking Group B5 Forecasting product demand for a retail chain to reduce cost of understocking and overstocking Group B5 Konpal Agrawal 61910895 Prakash Sarangi 61910902 Rahul Anand 61910361 Raj Mukul Dave 61910269 Ramchander

More information

FORECASTING ANALYSIS OF CONSUMER GOODS DEMAND USING NEURAL NETWORKS AND ARIMA

FORECASTING ANALYSIS OF CONSUMER GOODS DEMAND USING NEURAL NETWORKS AND ARIMA International Journal of Technology (2015) 5: 872-880 ISSN 2086-9614 IJTech 2015 FORECASTING ANALYSIS OF CONSUMER GOODS DEMAND USING NEURAL NETWORKS AND ARIMA Arian Dhini 1*, Isti Surjandari 1, Muhammad

More information

Introduction to Research

Introduction to Research Introduction to Research Arun K. Tangirala Arun K. Tangirala, IIT Madras Introduction to Research 1 Objectives To learn the following: I What is data analysis? I Types of analyses I Different types of

More information

Cluster-based Forecasting for Laboratory samples

Cluster-based Forecasting for Laboratory samples Cluster-based Forecasting for Laboratory samples Research paper Business Analytics Manoj Ashvin Jayaraj Vrije Universiteit Amsterdam Faculty of Science Business Analytics De Boelelaan 1081a 1081 HV Amsterdam

More information

Inventory Management for the Reduction of Material Shortage Problem for Pasteurized Sugarcane Juice: The Case of a Beverage Company

Inventory Management for the Reduction of Material Shortage Problem for Pasteurized Sugarcane Juice: The Case of a Beverage Company Inventory Management for the Reduction of Material Shortage Problem for Pasteurized Sugarcane Juice: The Case of a Beverage Company Roongrat Pisuchpen Faculty of Engineering, Industrial Engineering, Kasetsart

More information

When Automatic Forecasting Doesn t Do the Job

When Automatic Forecasting Doesn t Do the Job When Automatic Forecasting Doesn t Do the Job Presented by Eric Stellwagen Vice President & Cofounder Business Forecast Systems, Inc. estellwagen@forecastpro.com Eric Stellwagen Vice President & Cofounder

More information

EnterpriseOne JDE5 Forecasting PeopleBook

EnterpriseOne JDE5 Forecasting PeopleBook EnterpriseOne JDE5 Forecasting PeopleBook May 2002 EnterpriseOne JDE5 Forecasting PeopleBook SKU JDE5EFC0502 Copyright 2003 PeopleSoft, Inc. All rights reserved. All material contained in this documentation

More information

An evaluation of simple forecasting model selection rules

An evaluation of simple forecasting model selection rules MPRA Munich Personal RePEc Archive An evaluation of simple forecasting model selection rules Robert Fildes and Fotios Petropoulos Lancaster University Management School April 2013 Online at http://mpra.ub.uni-muenchen.de/51772/

More information

Inter American University of Puerto Rico Metropolitan Campus School of Management B.B.A. Program in Operations Management.

Inter American University of Puerto Rico Metropolitan Campus School of Management B.B.A. Program in Operations Management. Inter American University of Puerto Rico Metropolitan Campus School of Management B.B.A. Program in Operations Management Course Syllabus I. GENERAL INFORMATION Course Title: Code and Number: Credit Hours:

More information

Short Term Demand Forecasting for the Integrated Electricity Market

Short Term Demand Forecasting for the Integrated Electricity Market Student Journal of Energy Research Volume 2 Number 1 Article 1 2017 Short Term Demand Forecasting for the Integrated Electricity Market Katie Kavanagh Dublin Institute of Technology, d15125143@mydit.ie

More information

Forecasting Cash Withdrawals in the ATM Network Using a Combined Model based on the Holt-Winters Method and Markov Chains

Forecasting Cash Withdrawals in the ATM Network Using a Combined Model based on the Holt-Winters Method and Markov Chains Forecasting Cash Withdrawals in the ATM Network Using a Combined Model based on the Holt-Winters Method and Markov Chains 1 Mikhail Aseev, 1 Sergei Nemeshaev, and 1 Alexander Nesterov 1 National Research

More information

On the exact calculation of the mean stock level in the base stock periodic review policy

On the exact calculation of the mean stock level in the base stock periodic review policy On the exact calculation of the mean stock level in the base stock periodic review policy Eugenia Babiloni, Manuel Cardós, Ester Guijarro Universitat Politècnica de València (SPAIN) mabagri@doe.upv.es;

More information

General notice: Other product names mentioned herein are used for identification purposes only and may be trademarks of their respective companies.

General notice: Other product names mentioned herein are used for identification purposes only and may be trademarks of their respective companies. i SPSS Trends 16.0 For more information about SPSS software products, please visit our Web site at http://www.spss.com or contact SPSS Inc. 233 South Wacker Drive, 11th Floor Chicago, IL 60606-6412 Tel:

More information

Audits of Grant or Contribution programs May 2001

Audits of Grant or Contribution programs May 2001 Office of the Auditor General of Canada Audits of Grant or Contribution programs May 2001 Table of Contents Page 1. Purpose and Introduction 1 1.1 Purpose 1 1.2 Applicability 1 1.3 Prerequisite 1 1.4

More information

SimulationofCementManufacturingProcessandDemandForecastingofCementIndustry

SimulationofCementManufacturingProcessandDemandForecastingofCementIndustry Global Journal of Researches in Engineering: G Industrial Engineering Volume 16 Issue 2 Version 1.0 Type: Double Blind Peer Reviewed International Research Journal Publisher: Global Journals Inc. (USA)

More information

A Decision Support Method for Investment Preference Evaluation *

A Decision Support Method for Investment Preference Evaluation * BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 6, No 1 Sofia 2006 A Decision Support Method for Investment Preference Evaluation * Ivan Popchev, Irina Radeva Institute of

More information

Choosing the Right Type of Forecasting Model: Introduction Statistics, Econometrics, and Forecasting Concept of Forecast Accuracy: Compared to What?

Choosing the Right Type of Forecasting Model: Introduction Statistics, Econometrics, and Forecasting Concept of Forecast Accuracy: Compared to What? Choosing the Right Type of Forecasting Model: Statistics, Econometrics, and Forecasting Concept of Forecast Accuracy: Compared to What? Structural Shifts in Parameters Model Misspecification Missing, Smoothed,

More information

Promotional Indices Application in Business Forecasting of Promotion Events

Promotional Indices Application in Business Forecasting of Promotion Events Promotional Indices Application in Business Forecasting of Promotion Events MICHAL PATAK, VLADIMIRA VLCKOVA Faculty of Chemical Technology, Department of Economy and Man. of Chemical and Food Industry

More information

Demand Forecasting in the Supply Chain The use of ForecastPro TRAC

Demand Forecasting in the Supply Chain The use of ForecastPro TRAC Demand Forecasting in the Supply Chain The use of ForecastPro TRAC Marco Arias Vargas Global EMBA INCAE Distribution & Warehousing Inventory Forecasting 2 Complexity in the Supply Chain Vendor DC Materials

More information

MANAGING THE SUPPLY AND DEMAND UNCERTAINTY IN ASSEMBLY SYSTEMS

MANAGING THE SUPPLY AND DEMAND UNCERTAINTY IN ASSEMBLY SYSTEMS MANAGING THE SUPPLY AND DEMAND UNCERTAINTY IN ASSEMBLY SYSTEMS Hongze Ma^ and Kesheng Wang^ ^Logistics, Turku School of Economics and Business Administration, Turku, Finland; E-mail: Hongze.Ma@tukkk.fi.

More information

University of Jordan Jordan University Business School (JUBS)

University of Jordan Jordan University Business School (JUBS) 1. Department Name: Business Management 2. Program Name: Bachelor of Business Administration 3. Program Code 4. Course Code and Title: 1601413 Operations and Production Management 5. Course credits: 3

More information

INVENTORY MANAGEMENT IN HIGH UNCERTAINTY ENVIRONMENT WITH MODEL REFERENCE CONTROL

INVENTORY MANAGEMENT IN HIGH UNCERTAINTY ENVIRONMENT WITH MODEL REFERENCE CONTROL INVENTORY MANAGEMENT IN HIGH UNCERTAINTY ENVIRONMENT WITH MODEL REFERENCE CONTROL Heikki Rasku Hannu Koivisto Institute of Automation and Control, Tampere University of Technology, P.O.Box 692, Tampere,

More information

New Clustering-based Forecasting Method for Disaggregated End-consumer Electricity Load Using Smart Grid Data

New Clustering-based Forecasting Method for Disaggregated End-consumer Electricity Load Using Smart Grid Data New Clustering-based Forecasting Method for Disaggregated End-consumer Electricity Load Using Smart Grid Data Peter Laurinec, and Mária Lucká 4..7 Slovak University of Technology in Bratislava Motivation

More information

Forecasting Construction Cost Index using Energy Price as an Explanatory Variable

Forecasting Construction Cost Index using Energy Price as an Explanatory Variable Forecasting Construction Cost Index using Energy Price as an Explanatory Variable Variations of ENR (Engineering News Record) Construction Cost Index (CCI) are problematic for cost estimation and bid preparation.

More information

There has been a lot of interest lately

There has been a lot of interest lately BENCHMARKING SALES FORECASTING PERFORMANCE MEASURES By Kenneth B. Kahn Although different companies use different measures to evaluate the performance of forecasts, mean absolute % error (MAPE) is the

More information

Improving forecast quality in practice

Improving forecast quality in practice Improving forecast quality in practice Robert Fildes (Lancaster Centre for Forecasting) Fotios Petropoulos (Cardiff Business School) Panel Discussion Agenda The Forecasting Process Dimensions of quality

More information

ScienceDirect. Guidelines for Applying Statistical Quality Control Method to Monitor Autocorrelated Prcoesses

ScienceDirect. Guidelines for Applying Statistical Quality Control Method to Monitor Autocorrelated Prcoesses Available online at www.sciencedirect.com ScienceDirect Procedia Engineering 69 ( 2014 ) 1449 1458 24th DAAAM International Symposium on Intelligent Manufacturing and Automation, 2013 Guidelines for Applying

More information

King Saud University College Of Business Administration Management Department Course: Operations Management

King Saud University College Of Business Administration Management Department Course: Operations Management King Saud University College Of Business Administration Management Department Course: Section: 35003 ادا 371 Second Semester - 1432/1433 Instructor: Dr. Abdullah M. Aldakhil Office Hours: Saturday & Sunday

More information

Operation Management Forecasting: Demand Characteristics

Operation Management Forecasting: Demand Characteristics Paper Coordinator Co-Principal Investigator Principal Investigator Development Team Prof.(Dr.) S.P. Bansal Vice Chancellor, Maharaja Agreshen University, Baddi, Solan, Himachal Pradesh, INDIA Co-Principal

More information

Sales and Operations Planning

Sales and Operations Planning Sales and Operations Planning Alessandro Anzalone, Ph.D. Hillsborough Community College, Brandon Campus 1. Purpose of Sales and Operation Planning 2. General Design of Sales and Operations Planning 3.

More information

Principles of Operations Management: Concepts and Applications Topic Outline Principles of Operations Planning (POP)

Principles of Operations Management: Concepts and Applications Topic Outline Principles of Operations Planning (POP) Principles of Operations Management: Concepts and Applications Topic Outline Principles of Operations Planning (POP) Session 1: Operation Management Foundations Describe how today s business trends are

More information

FORECASTING AND DEMAND MANAGEMENT

FORECASTING AND DEMAND MANAGEMENT FORBUS JUNE 2013 EXAMINATION DATE: 7 JUNE 2013 TIME: 09H00 11H00 TOTAL: 100 MARKS DURATION: 2 HOURS PASS MARK: 40% (XN-88) FORECASTING AND DEMAND MANAGEMENT THIS EXAMINATION PAPER CONSISTS OF 3 SECTIONS:

More information

Application of SAS in Product Testing in a Retail Business

Application of SAS in Product Testing in a Retail Business Application of SAS in Product Testing in a Retail Business Rick Chambers, Steven X. Yan, Shirley Liu Customer Analytics, Zale Corporation, Irving, Texas Abstract: Testing new products is an important and

More information

Vipul Mehra December 22, 2017

Vipul Mehra December 22, 2017 Forecasting USD to INR foreign exchange rate using Time Series Analysis techniques like HoltWinters Simple Exponential Smoothing, ARIMA and Neural Networks Vipul Mehra December 22, 2017 Abstract Forecasting

More information

Better ACF and PACF plots, but no optimal linear prediction

Better ACF and PACF plots, but no optimal linear prediction Electronic Journal of Statistics Vol. 0 (0000) ISSN: 1935-7524 DOI: 10.1214/154957804100000000 Better ACF and PACF plots, but no optimal linear prediction Rob J Hyndman Department of Econometrics & Business

More information

OPTIMIZATION AND OPERATIONS RESEARCH Vol. IV - Inventory Models - Waldmann K.-H

OPTIMIZATION AND OPERATIONS RESEARCH Vol. IV - Inventory Models - Waldmann K.-H INVENTORY MODELS Waldmann K.-H. Universität Karlsruhe, Germany Keywords: inventory control, periodic review, continuous review, economic order quantity, (s, S) policy, multi-level inventory systems, Markov

More information

Winsor Approach in Regression Analysis. with Outlier

Winsor Approach in Regression Analysis. with Outlier Applied Mathematical Sciences, Vol. 11, 2017, no. 41, 2031-2046 HIKARI Ltd, www.m-hikari.com https://doi.org/10.12988/ams.2017.76214 Winsor Approach in Regression Analysis with Outlier Murih Pusparum Qasa

More information

BA3352 : Midterm on 8 October 2001

BA3352 : Midterm on 8 October 2001 BA3352 : Midterm on 8 October 2001 This is a closed textbook and lecture notes exam. You may not use a calculator so leave quantities as fractions, additions or products. Do not forget to define any variables

More information

Chapter 14. Simulation Modeling. Learning Objectives. After completing this chapter, students will be able to:

Chapter 14. Simulation Modeling. Learning Objectives. After completing this chapter, students will be able to: Chapter 14 Simulation Modeling To accompany Quantitative Analysis for Management, Eleventh Edition, by Render, Stair, and Hanna Power Point slides created by Brian Peterson Learning Objectives After completing

More information

NOWIcob A tool for reducing the maintenance costs of offshore wind farms

NOWIcob A tool for reducing the maintenance costs of offshore wind farms Available online at www.sciencedirect.com ScienceDirect Energy Procedia 35 (2013 ) 177 186 DeepWind'2013, 24-25 January, Trondheim, Norway NOWIcob A tool for reducing the maintenance costs of offshore

More information

STEPHEN CARSTENS RCBM (Pty) Ltd ABSTRACT

STEPHEN CARSTENS RCBM (Pty) Ltd ABSTRACT INCREASING THE COMPETITIVENESS OF MAINTENANCE CONTRACT RATES BY USING AN ALTERNATIVE METHODOLOGY FOR THE CALCULATION OF AVERAGE VEHICLE MAINTENANCE COSTS STEPHEN CARSTENS stephcar@global.co.za RCBM (Pty)

More information

PLANNING FOR PRODUCTION

PLANNING FOR PRODUCTION PLANNING FOR PRODUCTION Forecasting Forecasting is the first major activity in the planning, involving careful study of past data and present scenario to estimate the occurence, timing or magnitude of

More information

Recoupling the forecasting and inventory management processes

Recoupling the forecasting and inventory management processes Recoupling the forecasting and inventory management processes Aris A. Syntetos Cardiff University Panalpina Chair in Manufacturing & Logistics Foresight Practitioner Conference: North Carolina State University,

More information

Exploring simulation applications in education and research

Exploring simulation applications in education and research 29-May-15 1 Exploring simulation applications in education and research Durk-Jouke van der Zee Faculty of Economics & Business / Operations Tecnomatix Digital Manufacturing Solutions Day 2 Dr. ir. D.J.

More information

Enhancement of Intermittent Demands in Forecasting for Spare Parts Industry

Enhancement of Intermittent Demands in Forecasting for Spare Parts Industry Indian Journal of Science and Technology, Vol 8(25), DOI: 10.17485/ijst/2015/v8i25/53374, October 2015 ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 Enhancement of Intermittent Demands in Forecasting

More information

Economic Analysis for Business and Strategic Decisions. The Fundamentals of Managerial Economics

Economic Analysis for Business and Strategic Decisions. The Fundamentals of Managerial Economics Economic Analysis for Business and Strategic Decisions Chapter 1: The Fundamentals of Managerial Economics 1. Define the concept of time value of money. 2. Recognize the difference between value maximization

More information

MIT SCALE RESEARCH REPORT

MIT SCALE RESEARCH REPORT MIT SCALE RESEARCH REPORT The MIT Global Supply Chain and Logistics Excellence (SCALE) Network is an international alliance of leading-edge research and education centers, dedicated to the development

More information

Line Balancing in the Hard Disk Drive Process Using Simulation Techniques

Line Balancing in the Hard Disk Drive Process Using Simulation Techniques Line Balancing in the Hard Disk Drive Process Using Simulation Techniques Teerapun Saeheaw, Nivit Charoenchai, and Wichai Chattinnawat Abstract Simulation model is an easy way to build up models to represent

More information

Line Balancing in the Hard Disk Drive Process Using Simulation Techniques

Line Balancing in the Hard Disk Drive Process Using Simulation Techniques Line Balancing in the Hard Disk Drive Process Using Simulation Techniques Teerapun Saeheaw, Nivit Charoenchai, and Wichai Chattinnawat Abstract Simulation model is an easy way to build up models to represent

More information