There are two general approaches to forecasting. the quantitative approach and qualitative approach. Two samples for each(TOTALLY 4 SAMPLES)with reference to a MEDIUM-SIZED MANUFACTURING ENTERPRISE.

Name:
Instructor:
Course:
Date:
Forecasting approaches
Introduction
Enterprise managers have the desire to identify possible future events for them to determine the appropriate action to be taken or have the desired course action designed well in advance before its implementation. Every aspect of the business needs some sort of forecasting to determine its current and future possible demands, internally and externally: finance needs to establish the appropriate funding required and time-frame for borrowing; accounting should identify the cost estimates and projected profits; marketing to determine business competition strategies; human resource to identify recruitment, interview, layoff strategies, and personnel development such as training and counselling. These are just some examples with which forecasting can help in management decision making because the efficiency of the action plan adopted depends on the accurate prediction of future events. Irrespective of whether the enterprise provides services, sells or manufactures products, forecasting is inherent because they are so established to meet particular client needs. A forecast, therefore, refers to the process of estimating future events through a systematic combination and forward ‘casting’ of data obtained from the past, and a good one is invaluable in aiding managers “plan the system” in the long-run and “planning the use of the system” which is often in short to the intermediate term. It is, therefore, belligerent that managers are well-knowledgeable on the forecasting techniques and how they affect the operations as this is a crucial element of the management process.

Figure 1: The Forecasting Process (From https://gbr.pepperdine.edu/2010/08/artificial-intelligence-techniques-enhance-business-forecasts/)
Approaches to Forecasting
A good forecast used by management in predicting future decisions needs to be reliable, accurate, timely, actionable, easily understood, and expressed in whole units (Martinovic & Vesna 527). A forecast incorporating these elements can only be made possible subject to the availability of data obtained when gathering relevant statistical information. Qualitative data allows for a description of trends and the causal factors influencing particular events or actions and is widely dependent on the planner’s subjective judgment (Kurzak 177). Quantitative data is the principal information source for many enterprises, relying on a quantifiable analysis of processes and phenomena. It is of these types of data, the primary elements of a forecast, which we have two approaches to forecasting, i.e. the qualitative and quantitative methods which can be used individually or combined to evolve the forecast.
Qualitative Approach
The qualitative approach to forecasting involves methods consisting of subjective inputs devoid of “precise numerical description” (Armstrong 372). This method permits the inclusion of soft information such as personal opinions and human factors which are often not factored in the quantitative approach because of their nature that makes it difficult or even impossible for their quantification. The techniques employed under this approach require the identification of people’s ideas and decisions related to the current and future situation (Kocaoglu, Acar, & Yilmaz 28). Expert opinions and judgments are treated on their experiences and subjective factors.
The advantages of this approach efficient inefficient or uncertain or is prone to change excessively over time during consideration. The data and information required in the formulation of the forecast can be obtained from multiple sources such as customers, managers, staff, sales people, or other experts from companies in the same industry. The approach gives allowance for the development of opinion diversity and the anonymity of the experts involved increases its reliability (Pilinkienė 21).
The disadvantages of this approach lie on the basis that, since they are dependent on subjective and abstract experiences relying on “intuition,” the results are likely to be biased towards the crafter’s point of view even though not informed by any subjective backing. There is no consideration for fluctuations in data sequence, assessment of interrelationship between economic data, and cannot be applied in developing short-term forecasts (Pilinkienė 21).
Delphi technique
The Delphi method is an example of a qualitative technique developed by the RAND Corporation in the US to predict “the hypothetical Soviet attempt to USA’S production capacity and their ability to maintain a war effort” (Cornel & Mirela 31). It is aimed at achieving consensus among experts on a particular research problem through an “interactive firms set of the carefully designed sequential questionnaire with a summarized feedback of views derived from earlier responses” (Cornel & Mirela 32). Once a survey focusing on a particular problem has been developed, the experts’ panel is identified, and the questionnaire administered to them. The questionnaires are answered independently, responses summarized, and a further questionnaire developed based on the findings and sent to the same experts’ panel. They rate and prioritize the ideas independently, and the process continues until when the investigators identify firm positions and agreement on the topic among the experts.
The advantages of this technique are that participants have adequate time to brainstorm their ideas to evolve better quality responses as well as creating records of the expert group’s responses that can be alluded to when needed; its flexibility ensures that it can be applied to diverse situations and multiple complex problems. Many experts across different geographical locations can be involved thus assuring the generation of quality opinions and there is adequate time to the experts to consider their outcomes and recommendations critically. However, the results so obtained are dependent on the participant’s quality of expertise and there is the possibility of the researcher influencing the result.

Figure 2: Schematic Representation of the Delphi Technique (From https://www.tools4management.com/article/achieve-expert-consensus-with-the-help-of-delphi-method/)
Use of Sales Personnel
Sales people and agents are the ones who are close to the consumers and can be utilized effectively in the accurate prediction of the market needs and preferences, as well as an evaluation of the market size and market share (Armstrong & Rod 98). The focus groups are often the sales agents and the customers, and the honest opinions can help evaluate the market needs and requirements. This includes conducting a client or sales people surveys, thus make a customized product-by-product forecast on the individual sales territory they occupy. This bottom-up approach enables the organization evaluate future sales with enhanced accuracy and thus evolve reasonable trends and plans for the future market. The data so obtained can be used to model future forecasts or complement the qualitative and quantitative data collected by the enterprise. It is a simple approach to comprehending and use, uses specialized knowledge of those people closest to the consumer and makes it possible to specifically concentrate on a particular demography of users by territory or product. However, most sales people are rarely experienced in forecasting approach, and thus they may give information that is overly pessimistic or optimistic on the inaccuracies and predictions because of the macro-micro economic factors beyond their scope (Smaros & Markus 142).
Quantitative Approach
This method is often objective and relies on mathematical techniques and less on the judgment during computation, and often involve an extension of historical data or developing associative models that utilize explanatory (causal) variables to make a forecast (Martinovic & Vesna 528). They mainly involve the analysis of objective or hard data primarily because historical data is widely availed, both from within the organization, businesses in the same industry, scholarly sources or even from government data. The process of gathering information for this approach aims at the description of a phenomenon to a large pool of participants to establish the demographics and relationship in each target market segment. They allow for the incorporation of models found in statistics, econometrics, mathematics, and ultimate optimization in assessing the overall relationship patterns (Gor 153). This makes it quite an invaluable approach because extrapolation or manipulation of the data can be undertaken in a manner that is in tandem with the overall business goals of the organization. This is an essential approach, particularly when doing short-term forecast as it can help evaluate the study group to the control group, in determining whether the firm’s strategies are being adequately met and linking the specific causal factors to the specific outcomes.
The advantage of this approach lies in the uncomplicated and easy access to data, a possibility to determine possible future business change points, and helps to assess a particular economic indicator’s fluctuation and interrelationships (Pilinkienė 21). It makes it possible to have information collected from many stakeholders, allows for comparisons and developing of causal relations between the demographic groups and variables under study, and provides informative content for initiating strategic goals or guidelines. The disadvantages include that it is not the appropriate method to forecast the demand for a new product, can often complicate the application and misinterpretation of results, and high costs are involved as it regularly requires continuous data compilation and market analysis (Pilinkienė 21). This procedure may also pose difficulties in recognition of new phenomena that may not be documented and requires careful analysis and interpretation where there is no control group.
Time-series Methods
Manufacturing companies often have enough historical consumption data for use in forecasting analysis (Gor 148). This data could relate to items as the demand placed on particular products, periodic sales trends of individual products, nature of consumers, consumer price index, and profits gained. Time-series methods are used to undertake the statistical analysis of this historical data to design future forecasts, following routine observations over regular time intervals (e.g. yearly, quarterly, monthly, weekly, daily, and hourly). It is assumed that consumer demand is static and as such past relationships will hold constant into the future, and the analysis of time-series data helps the analyst recognize the inherent characteristics of the series. It can in the form of trends which relate the downward or upward movements of data in the long-term such as to include changes in commodity price and relative demand, often influenced by factors such as changing incomes, and cultural and population shifts. Seasonal variations relate the constant periodic fluctuations in the series occurring due to changing consumer patterns and social habits, such as the demand for cold sodas in the hot weather or coffee in the cold season; or alcoholic beverages during the holiday seasons. Cyclical variations are wavelike and may occur for periods of more than a year, and are related to political, economic, and agricultural that may alter the production of the raw materials or effective distribution or supply of the product in the market.

Figure 3: Example of Linear Progression time-series (From https://www.isixsigma.com/tools-templates/graphical-analysis-charts/making-sense-time-series-forecasting/)
Market Research
There are many experienced and competent market research firms that have specialized in the skill that can be hired by a firm to conduct this forecasting method. This is an excellent method that can be employed especially when a manufacturing company is seeking to introduce a new product it has developed or seeking ideas on a preferred product type. The contracted firms can employ questionnaires, or telephone interviews and other means to collect information from consumers of clearly defined demographics about items such as product preference, views on the existing products (more so what competitors have to offer), customer habits, and their incomes (purchasing power) (Smith & Gerald 75). There is an increased likelihood of the information so obtained by the research firm to be customized and specialized to the needs of the manufacturing company on the particular line of action to pursue to satisfy the clients fully with the manufactured product, with often concentrations in elements such as testing a brand name, customer care relations, and the development of a new product (Ogbadu 14). The firms adopt a standard approach in the process and thus the results can be easily compared to the industry to identify evolutionary concepts.
The numerical data obtained for a small-sized medium enterprise as a supermarket would take advantage of the availability of point of sale data, barcode technology, and the clients, utilizing this information in obtaining precise information. Such information can then be saved onto a database employing customer relationship management software, sharing this with others in the field using enterprise resource planning systems software. The evolving technologies have made it possible for such business enterprises to register all information regarding a purchase such as a customer identity and quantities purchased as well as an indication of customer loyalty, e.g. using customer loyalty cards and the client’s product portfolio.
Conclusion
In an ever increasingly competitive economic environment, managers need to continuously stay afloat on forecasting approaches that can help them predict future trends and take a reactive approach (meet the demand) or a proactive approach (actively influencing the demand). Many techniques are available broadly falling under the qualitative, and quantitative methods and none can work at best in all situations. The technique to be adopted would depend on the accuracy needed with consideration of the cost implications, availability of historical data, the time required for analysis, type of forecast needed (short-term or long-term), and the ability of decision makers to make use of particular techniques efficiently. For manufacturing companies, management can rely on technological development and use both approaches in developing a new product, customizing the existing product to meet customer needs, or adjusting prices to handle competition. Forecasting techniques are thus a critical planning tool that managers must adopt if they are to navigate the often murky and competitive business environment efficiently.
Works Cited
Armstrong, J. Scott. “Selecting forecasting methods.” Principles of forecasting. Springer US, 2001. 365-386.
Armstrong, J. Scott, and Rod Brodie. “Forecasting for marketing.” (1999): 92-119.
Cornel, Lazăr and Mirela, Lazăr. Delphi – The Highest Qualitative Forecast Method. Petroleum-Gas University of Ploieşti. Vol. LX No. 1/2008: 31-36
Gor, Ravi Mahendra. “Forecasting Techniques.” Industrial Statistics and Operational Management. (2009): 142-163.
Kocaoglu, Batuhan, Acar Zafer A. & Yilmaz, Behlül. Demand Forecast, Up-to-date Models, and Suggestions for Improvement: An Example of a Business. Journal of Global Strategic Management. 8.1 (2014): 26-37.
Kurzak, Lucjan. “Importance Of Forecasting In Enterprise Management.” Advanced Logistic systems 6.1 (2012): 173-182.
Martinovic, Jelena, and Vesna, Damnjanovic. “The sales forecasting techniques.” Materials of International scientific days” Competitiveness in the EU–challenge for the v4 countries (2006): 526-531.
Ogbadu, Eligah E. “Appraisal of the Practical Application of Marketing Research by SMEs in Nigeria.” International Journal of Academic Research in Business and Social Sciences 3.2 (2013): 10.
Pilinkienė, Vaida. “Selection of market demand forecast methods: Criteria and application.” Engineering Economics 58.3 (2015).
Småros, Johanna, and Markus Hellström. “Using the assortment forecasting method to enable sales force involvement in forecasting: a case study.” International Journal of Physical Distribution & Logistics Management 34.2 (2004): 140-157.
Smith, Scott M and Gerald S Albaum. Basic Marketing Research. 1st ed., Qualtrics Labs, 2012.