Introduction
When it comes to any business operations, a long-term strategy and predictions play a significant role in the business’s success. Forecasting, for its past, has become an integral part of the business strategy and tactics. The procedure implies the complex process of gathering historical data in order to generate predictions for the future of the operations. According to Taylor and Letham (2018), forecasting techniques allows business to “engage in capacity planning to efficiently allocate scarce resources and goal setting in order to measure performance relative to a baseline” (p.2). As a result, the future of the company’s operations can be objectively based on the expectations set by the forecast.
However, when it comes to forecasting, relying solely on the average estimates could compromise the accuracy of the findings, as there are various aspects that might affect the data at a certain point in time. For example, when estimating the demand data for the forecast, the indicators should be deseasonalized or presented in a way unaffected by seasonal fluctuations and other potential biases. Moreover, it is necessary to calculate the periodicity of the data in order to make sure that the seasonal fluctuations are repeated over a certain period of time. The aim of this paper is to apply the forecasting techniques to a case study and discuss the implications of forecasting to operations.
Exponential Smoothing Forecasting Calculation
One of the most commonly accepted ways to employ quantitative forecasting is exponential smoothing. This phenomenon stands for the forecasting using time series with a certain systematic and seasonal component. Thus, by adding the alpha component to the calculation formula, the weight of the calculations can be lightly or heavily dependent on such external factors as seasonal fluctuations. Hence, in the formula below, given the demand for the past periods, the forecast for the next period using different alpha indicators can be done:
Ft+1 = α yt + (1-α) Ft
In this case, the forecasts for the latest period are based on how smoothing value affects older period forecasts. In the table below, this calculation is performed for the given demand over twelve periods:
Table 1. Single Exponential Smoothing Forecast.
Given the data above, it becomes evident that both forecasts demonstrate the prediction of demand in a progression, estimating no potential downfalls. However, in period five demand, a significant drop may be observed, which was likely to come as unprecedented for the business operations. Hence, it is reasonable to assume that both forecasting approaches are somehow distanced from the objective data, as they could not predict the demand’s trajectory.
However, when comparing the two alpha measures, the 0.8 forecast coefficient is more reliable. In the case scenario of 0.2 coefficient, the demand demonstrates findings far from the actual numbers. Even though it implies growth in the thirteenth period, the difference between the values is significant. On the other hand, the forecasting implications demonstrated with a 0.8 alpha coefficient represent values closer to the truth, as they place more emphasis on the recent data. Indeed, in a formula with a 0.8 coefficient, this value is given to the latest period data, whereas the rest, 0.2 value, is placed on the old one. For this reason, it can be concluded that when a higher value is given to the alpha coefficient, the quantitative outcomes are likely to be more precise in the short term. On the other hand, smaller values are designed to create long-term calculations of growth with less accurate quantitative predictions. Thus, the reliability of these calculations depends on the forecasting intentions, as low values are more relevant in the long-term anticipation, whereas the short-term accuracy of high-value calculations is better.
Alternative Calculation Method
Since the model above present no trend or seasonality, and the only known value include the number of periods and demand, another means besides simple exponential smoothing is the moving average method. The moving average method implies the estimation of how the mean moves over a certain periodicity. For example, the given 12 periods of the data can be perceived as one fiscal year divided into four quarters. Thus, it is possible to calculate the mean for each quarter and define the moving average for each year to estimate the trend and produce long-term forecasting. Undeniably, this approach to forecasting has a number of disadvantages, including the disregard of the data complexity and fluctuation. Hence, when employing this method, the forecast should be perceived as a basis for a future, more detail-oriented analysis.
Operations Planning
Once the demand forecasting is completed, the next step for the business operations is to create an extensive operations planning that includes the required materials, workforce, and costs for meeting the demand for the product in the long term. Essentially, there are three types of operations planning: chase, level, and mixed strategies. While all these strategies can be embraced by the companies, they have distinct differences.
A chase operations strategy stands for the process of planning manufacturing in a way that adjusts the workforce exactly to meet the demand for a specific product. Such a strategy works well for companies with a narrow specialization that work on a make-to-order basis. A make-to-order basis is a business phenomenon that allows the manufacturers to customize the product directly according to the customer demand, minimizing the intermediaries. Hence, such an approach will be disadvantageous for large companies that seek consistency and large-scale manufacturing.
For such enterprises, adopting a level strategy is a more beneficial choice. In level operations planning, the emphasis is placed on the safety stocks and inventory fluctuations, and backlog levels. In such a way, companies will mass production, and a stable workforce is capable of tracking the potential changes in the demand production depending on logistics, seasonal fluctuations, and other external aspects. According to Rakićević (2018), level strategy is “achieved through two directions, “maintain a level workforce,” and maintain a steady output rate” (p. 1069). In order to meet these expectations, the company can find itself at a disadvantage of allocating high costs on constant inventory.
Finally, a mixed approach to operations planning is suitable for the firms willing to combine both make-to-order and make-to-stock operations by meeting short-term demands and planning long-term level production. The evident advantage of such an approach would be the flexibility of production forecasting and planning. On the other hand, a mixed production strategy entails a complex hierarchy of operations management. For this reason, the strategic planning for the firm should include a professional team to operate both staff and contractors. According to a scholarly review by Kristensen and Jonsson (2019), there is no evidence that the choice of the manufacturing strategy explicitly affects performance. However, it is still necessary to account for the size of the firm and the type of products manufactured when adopting one of the planning strategies.
Planning Systems
Another significant part of securing an efficient allocation of resources and productivity tracking is the implementation of Enterprise Resource Planning (ERP). ERP is essentially a system that incorporates the data from various firms’ departments to create a single interconnected database that presents insights into the performance (Chofrethh et al., 2020). According to Sutduean et al. (2019), ERP plays a significant role in the efficient management of the supply chain integration, especially when supply chain is located internationally. The obvious advantages of such a system are efficiency, instant access to converted data, and better planning opportunities. The research demonstrates that implementing ERP is a step towards successful organizational performance, as the integration of various data types helps draw prospects for the firm (Sutduean et al., 2019). Conversely, the drawbacks include costs, complexity, and the need for staff training before the implementation.
Thus, when considering the implementation of the ERP, the company should make sure that they are capable of providing enough time, money, and workforce to integrate the framework. Indeed, for the system to work, a company needs to secure a commitment from every department, training the employees on working with data in common software. A prime example of successful ERP implementation is Nestle, as it managed to implement ERP across the supply chain worldwide, eliminating outdated databases (Carlton, 2021). Hence, a corporation should implement this planning system, given that enough labor and financial resources are provided.
Reflection
When paying attention to the specifics of forecasting and its calculation, it has become evident that companies lose effort and money when having no interest in the long-term implications of the product demand. The process of forecasting, especially when the product is closely correlated with external factors such as social trends or seasonal fluctuation, provides the company with an opportunity to choose an efficient operations model that would realistically meet the product demand and suggest a working employment pattern for the firm. The key component of successful forecasting is to connect the obtained data with the planning operations and a framework of resource allocation, as being aware solely of the demand trends may become a liability for the firm.
However, it has also been noted in the process of learning that the forecasting itself can be rather subjective, especially for a new company with little historical data available. Thus, analytics should be a direction and a tool for planning rather than a full-scale margin that dictates the behavior of both the client and manufacturer. Later, this tool should be supported by a result-oriented financing and planning agenda. Hence, it can be concluded that the modern context of electronic data storage and computing makes forecasting an essential ground tool for planning and allocating resources for operations.
References
Carlton, R. (2021). Four ERP implementation case studies you can learn from. ERPFocus. Web.
Chofrethh, A. G., Goni, F. A., Klemeš, J. J., Malik, M. N., & Khan, H. H. (2020). Development of guidelines for the implementation of sustainable enterprise resource planning systems. Journal of Cleaner Production, 244. Web.
Kristensen, J., & Jonsson, P. (2018). Context-based sales and operations planning (S&OP) research: A literature review and future agenda. International Journal of Physical Distribution & Logistics Management, 48(1), 19-46. Web.
Rakićević, Z. (2018). The S&OP: Practical and advanced mid-term production planning. Doing Business in the Digital Age: Challenges, Approaches, and Solutions (IX International Symposium SymOrg 2018, pp. 1066-1073).
Sutduean, J., Singsa, A., Sriyakul, T., & Jermsittiparsert, K. (2019). Supply chain integration, enterprise resource planning, and organizational performance: The enterprise resource planning implementation approach. Journal of Computational and Theoretical Nanoscience, 16(7), 2975-2981. Web.
Taylor, S. J., & Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1), 37-45. Web.