Home

Articles Tools About Contact

Username: * Password: * Remember me   Create new account I forgot my password

Home > Articles > Browse Articles > Article

The increased competition and sophistication of consumer demand has forced companies to optimize operations. The company’s agility in response to consumer demand is key to its success, and the ability to predict the demand is critical for effective supply chain management. This article is the first in a series of four articles that outline popular techniques of forecasting demand, designed to meet the needs of both the student and the logistics professional. Purchasesmarte.com also provides some free online hands-on tools and utilities for experiencing these techniques.

        1.  Introduction: importance of demand predictions

With the increase in competition, increasingly sophisticated consumer demand, and continuously changing environment, companies are forced to improve their operations. The company's agility in response to demand has been identified as a key competitive strength, requiring aggressive focusing on supply chain management [1].
In order to properly manage customer needs, forecasting demand is invaluable, and is a key factor affecting success [2]. Obtaining information, such as demand and price, has always been a major commercial objective. Demand information affects production scheduling, inventory control, and delivery plans, while price information affects buyer allocation of purchasing quantities, which in turn affects demand. Demand information is therefore a key factor in supply chain management, with the objective of better matching supply and demand to reduce costs of inventory and stockout, while distorted demand and price information may cause supply and demand mismatches [2].

        2.  Inventory control and forecasting approaches

Good knowledge about demand enables vendors to maintain minimal inventory levels. Therefore, demand forecasting is a key aspect of inventory control. There are two basic forecasting approaches: assessing future market requirements, and using demand history [3].
Assessing demand requires knowledge about customers, products, and some background conditions that are evaluated by customer surveys and market studies. One method of determining customer demand is assessing potential sales; conversion factors calculated as an expected average demand by item can be a good estimate. Historical forecasting is a basic tool for inventory control that is best used while monitoring environment changes that may require readjustment for result accuracy. Historical forecasting methods are based on mathematical manipulation of historical data [3].
One of the most simple and popular techniques of historical forecasting, is the moving average.

        3.  The moving average forecasting technique

This is a common forecasting method because it’s simple to apply and understand. It is most successful where the demand fluctuates widely, making it difficult for sophisticated methods to identify demand trends, but can be as easily used when demand exhibits a basic pattern [4]. It is popular for determining market trend changes [5] and is commonly used as a basis for more sophisticated techniques [6]. The shortcoming of the technique is that it doesn’t respond well to changes, it attaches equal importance to all periods, and requires a large amount of data to calculate the average each time [2].
This is performed by calculating the average demand for a constant number of past observations. Example: if we have the following history of demand of 5 months, and define the moving average for 3 months:

Table 1: Example for 5-month demand

Months	1	2	3	4	5	6
Demand	11	20	14	12	19	?

* Editor’s note: use the free online moving average generator to generate forecasts using the moving average technique.

Figure 1: Graphical representation of the demand example


*Editor's Note: you can generate your own quick charts online here.

As the figures show, it’s not visually possible to identify a trend that will enable the determination of a “thumb rule”. According to the moving average, the demand forecast for the 6th month will be the average of the last 3 months: (14+12+19) / 3 =  15.  The demand for the 7th month will be the average of the last three, which includes the demand for the 6th month, as follows: (12+19+15) = 15.3. This way, it’s possible to predict the demand of future periods.

Table 2: An example for predicting the demand for the last month in a one-year period, with a moving average of 3 observations.

Month	Demand (Thousands)	Prediction: Moving Average of 3 previous observations
1	18
2	23
3	33
4	27	24.67
5	21	27.67
6	26	27.00
7	44	24.67
8	32	30.33
9	21	34.00
10	35	32.33
11	36	29.33
12	?	30.67

The demand forecast for the 4th month will be the average of the last 3 months: (18+23+33) / 3 = 24.67.  The demand for the 5th month will be the average of the last three: (23+33+27) / 3 = 27.67 , and so on. This way, the demand for the 12th month will be the average of the last three.
Selecting the best value for the number of past observations can be done with the MAD technique.

        4.  The weighted moving average forecasting technique

This method places a larger weight on recent values and less on others, because values that are more recent are probably more predictive.  The technique is applied by multiplying the previous observation by a weight [7].

*Editor’s note: use the free online moving average generator that enables weighted moving average forecasts.

The model's shortcomings are [7]: as the number of observations increases, the model smoothes fluctuations, making it less sensitive to changes: since it is an average, it won’t pick up trends; and it requires a large dataset. Table 3 shows an example of a weighted moving average of 3 observations, in order to predict the demand for the last month.

Table 3: An example of weighted moving average

Month	Demand (Thousands)	Weight	Demand × Weight	Prediction: Moving Average of 3 previous observations
1	18	0.9	16.2
2	23	0.91	20.93
3	33	0.92	30.36
4	27	0.93	25.11	22.50
5	21	0.94	19.74	25.47
6	26	0.95	24.7	25.07
7	44	0.96	42.24	23.18
8	32	0.97	31.04	28.89
9	21	0.98	20.58	32.66
10	35	0.99	34.65	31.29
11	36	1	36	28.76
12	?			30.41

The last month’s observation is given a weight of 1, and the weight for each month before it is decreased by 0.01. The demand forecast for the 4th month will be the average of the last 3 [demand × weight] products: (16.2+20.93+30.36) / 3 = 24.67.  This way, the demand for the 12th month will be the average of the last three [demand * weight] products.
Selecting the best value for the number of past observations can be done using the MAD technique.

        5.  Summary

Increased high competition forces companies to improve their operations and optimize supply chain management. Predicting customer demand is a critical part of the supply chain and a key factor in the company’s success. This article is the first in a series of four articles that demonstrate popular techniques of demand forecasting. In this article, two simple popular techniques have been demonstrated: weighted and unweighted moving average.

Part 2 describes the simple exponential smoothing technique
Part 3 presents regression analysis
Part 4 discusses selection among all techniques

        6. Final words on implementing processes and leadership
Moving Average is a very common technique, which can be implemented and combined with other forecasting techniques. The advance of technology, which supports companies innovation [8], allows the implementing of new strategies and tactics [9, 10], however, these depends not only on efficient processes [11], but also on leadership and management [12-14], which lay the ground for team work and social interactions [15], affecting even employee’s happiness [16], and eventually even the company’s stock price [17].

Bibliography

1. Changchien, S.W. and H.-Y. Shen., Supply chain reengineering using a core process analysis matrix and object-oriented simulation. Information & Management, 2002. 39(5): p. 345-358.
2. Sethi, S.P., H. Yan, and H. Zhang, Inventory and Supply Chain Management with Forecast Updates. . 2005: Chapter 1. Springer
3. Wild, T., Best practices in inventory management 1997: Chapter 10. Wiley.
4. Wikner, J., Analysis of smoothing techniques: application to production-inventory systems. Kybernetes, 2006. 35(9): p. 1323-1347.
5. Tse, R.Y.C., An application of the ARIMA model to real-estate prices in Hong Kong. Journal of Property Finance, 1997. 8(2): p. 152-163.
6. Loh, E.Y.L., Profiting from moving averages and time-series forecasts: Asian-pacific evidence. Asia Pacific Journal of Economics & Business, 2005. 9(1): p. 62-82,89.
7. Aghazadeh, S.-M., Revenue Forecasting models for hotel management. The Journal of Business Forecasting, 2007. 26(3): p. 33-37.
8. Eckhaus, E., Towards tourism business change . Review of International Comparative Management / Revista De Management Comparat International, 2017. 18(3): p. 274-286.
9. Eckhaus, E., G. Klein, and J. Kantor, Experiential learning in management education . Business, Management and Education, 2017. 15(1): p. 42-56.
10. Eckhaus, E., Barter trade exchange to the rescue of the tourism and travel industry . Journal of Shipping and Ocean Engineering, 2011. 1(2): p. 133-140.
11. Eckhaus, E., K. Kogan, and Y. Pearlman, Enhancing strategic supply decisions by estimating suppliers’ marginal costs . Journal of Supply Chain Management 2013. 49(4): p. 96-107
12. Eckhaus, E., A shift in leadership . Academy of Strategic Management Journal, 2017. 16(1): p. 19-31.
13. Eckhaus, E. and Z. Sheaffer, Managerial hubris detection: the case of Enron . Risk Management, 2018: p. Published online at https://link.springer.com/article/10.1057/s41283-018-0037-0.
14. Klein, G. and E. Eckhaus, Sensemaking and sensegiving as predicting organizational crisis . Risk Management, 2017. 19(3): p. 225-244.
15. Ben-Hador, B. and E. Eckhaus, The different impact of personal social capital and intra-organizational SC: The Enron case study . International Journal of Organization Theory & Behavior, 2018. 21(1): p. 28-47.
16. Eckhaus, E., Measurement of organizational happiness , In J. Kantola, T. Barath, & S. Nazir (Eds.), Advances in Human Factors, Business Management and Leadership. AHFE 2017. Advances in Intelligent Systems and Computing (Vol. 594, pp. 266-278). Cham: Springer International Publishing.
17. Eckhaus, E., Corporate transformational leadership's effect on financial performance . Journal of Leadership, Accountability and Ethics, 2016. 13(1): p. 90-102.

copyright © Purchasesmarter.com. All rights reserved. The material may not be published, rewritten, broadcast, or redistributed. Any reproduction in whole or part by and individuals or organizations will be held liable for copyright infringement to the full extent of the law. 

---