CHAPTER-3-FORECASTINg a lecture nottes for ttthe stud

CHAPTER-3-FORECASTINg a lecture nottes for ttthe stud

1. Chapter 3: forecasting

2. Definition of forecast Forecasts are a basic input in the decision processes of operations management because they provide information on future demand.

3. Importance of forecasting to operations management The primary goal of operations management is to match supply to demand. A forecast of demand is essential for determining how much capacity or supply will be needed to meet demand. Operations need to know what capacity will be needed to make staffing and equipment decisions, budgets must be prepared, purchasing needs information for ordering from suppliers, and supply chain partners need to make their plans.

4. Aspects of forecasts LEVEL OF DEMAND Can be a function of some structural variation, such as a trend or seasonal variation. DEGREE OF ACCURACY Is a function of the ability of forecasters to correctly model demand, random variation, and sometimes unforeseen events. Example: Forecasting of Sales of Business. Using Data Models for accuracy.

5. Forecasts are the basis of: Budgeting Planning Capacity Sales Production Inventory Purchasing

6. Examples of uses of forecasts in business organization Accounting- New product/process cost estimates, profit projections, cash management. Finance- Equipment/equipment replacement needs, timing and amount of funding/borrowing needs. Human resources- Hiring activities, including recruitment, interviewing, and training layoff planning, including outplacement counseling. Marketing- Pricing and promotion, e-business strategies, global competition strategies. Operations. Schedules, capacity planning, work assignments and workloads, inventory planning, make-or-buy decisions, outsourcing, project management. Product/service design- Revision of current features, design of new products or services.

7. Uses of forecasts to high level management PLAN THE SYSTEM Planning the system generally involves long-range plans about the types of products and services to offer, what facilities and equipment to have, where to locate, and so on. PLANNING THE USE OF SYSTEM refers to short-range and intermediate-range planning, which involve tasks such as planning inventory and workforce levels, planning purchasing and production, budgeting, and scheduling.

8. Forecasting techniques and their common features Forecasting Assumes Continuity of the Causal System Forecasting methods are based on the assumption that the underlying causal system—i.e., the factors that influence past trends—will remain stable in the future. This assumption means that if certain conditions influenced demand in the past, these conditions will continue to do so in the future. Example: Consider a retail company that uses historical sales data to forecast future demand for winter jackets. The forecasting model might assume that the relationship between cold weather and increased jacket sales will hold true in the future. However, if an unanticipated weather pattern occurs—such as an unusually warm winter—the forecast may be inaccurate. Similarly, if a competitor introduces a new, highly desirable jacket model or if there’s a sudden economic downturn, these unplanned occurrences can disrupt the forecasted trends.

9. Importance: Managers must understand that forecasts are not infallible. They need to stay informed about changes in the external environment (e.g., weather, economic conditions) and be prepared to adjust their forecasts or make strategic decisions based on new information. This vigilance helps in mitigating the risk of relying solely on outdated or overly rigid models.

10. II. Forecasts are Not Perfect; Allowances for Errors - Forecasting is inherently imperfect due to the presence of randomness and uncertainty. Forecasts provide estimates based on past data, but they cannot account for all possible variables and unforeseen events. Therefore, it's essential to recognize and accommodate forecast errors. Example: A manufacturing firm might forecast that it will need 10,000 units of a specific component for the next quarter based on past usage patterns. However, due to unexpected changes in customer preferences or supply chain disruptions, the actual requirement might turn out to be 12,000 units or only 8,000 units. The difference between the forecasted and actual demand represents forecast error. Importance: Managers should build flexibility into their planning processes to handle such uncertainties. This can be achieved by maintaining safety stock, using buffer resources, or regularly updating forecasts with new data. Being aware of the limitations of forecasts helps in planning for variability and reduces the impact of unexpected deviations.

11. III. Group Forecasts Tend to be More Accurate than Individual Forecasts - When forecasting for a group of items or aggregate demand, the overall accuracy tends to be higher compared to forecasting for individual items. This is because errors in individual forecasts may cancel each other out when aggregated, leading to a more stable and accurate overall forecast. Example: A company forecasts demand for various types of office supplies, including pens, notebooks, and staplers. Forecasting the total demand for all these supplies together is generally more accurate than forecasting the demand for each item individually. This is because fluctuations in the demand for one type of supply might be offset by fluctuations in another, leading to a more balanced and accurate aggregate forecast. Importance: By grouping items or aggregating forecasts, organizations can reduce the impact of variability and errors associated with individual forecasts. This approach helps in better inventory management and resource planning, leading to more efficient operations.

12. IV. Forecast Accuracy Decreases with Time Horizon The accuracy of forecasts typically decreases as the forecast horizon extends further into the future. Short-term forecasts generally face fewer uncertainties compared to long-term forecasts, which are more prone to unpredictable changes and variations. Example: A company may be able to forecast monthly sales for the next three months with high accuracy due to stable short-term conditions. However, forecasting sales for the next three years would involve more uncertainties, such as market trends, technological advancements, and economic shifts, making the forecast less reliable. Importance: Organizations that can adapt quickly to changes in demand and have a shorter forecasting horizon benefit from higher forecast accuracy. This is particularly relevant for industries with fast-changing conditions, such as technology and fashion. Flexible businesses can use short-range forecasts to make timely decisions and adjust their strategies as needed, while less flexible organizations might face challenges with longer-range forecasts and must prepare for potential inaccuracies.

13. Elements of a good forecast The forecast should be TIMELY. Usually, a certain amount of time is needed to respond to the information contained in a forecast. For example, capacity cannot be expanded overnight, nor can inventory levels be changed immediately. Hence, the forecasting horizon must cover the time necessary to implement possible changes. The forecast should be ACCURATE and the degree of accuracy should be stated. This will enable users to plan for possible errors and will provide a basis for comparing alternative forecasts. The forecast should be RELIABLE ; it should work consistently. A technique that sometimes provides a good forecast and sometimes a poor one will leave users with the uneasy feeling that they may get burned every time a new forecast is issued.

14. The forecast should be expressed in MEANINGFUL UNITS . Financial planners need to know how many dollars will be needed, production planners need to know how many units will be needed, and schedulers need to know what machines and skills will be required. The choice of units depends on user needs. The forecast should be in WRITING . Although this will not guarantee that all concerned are using the same information, it will at least increase the likelihood of it. In addition, a written forecast will permit an objective basis for evaluating the forecast once actual results are in. The forecasting technique should be SIMPLE TO UNDERSTAND AND USE . Users often lack confidence in forecasts based on sophisticated techniques; they do not understand either the circumstances in which the techniques are appropriate or the limitations of the techniques. Misuse of techniques is an obvious consequence. Not surprisingly, fairly simple forecasting techniques enjoy widespread popularity because users are more comfortable working with them. The forecast should be COST-EFFECTIVE : The benefits should outweigh the costs.

15. FORECASTING AND THE SUPPLY CHAIN Accurate forecasts are crucial in supply chain management. Inaccurate forecasts can cause both shortages and excesses, leading to missed deliveries, work disruptions, and poor customer service. Shortages can disrupt operations, while overly optimistic forecasts can result in excess inventory and increased costs. Both scenarios negatively impact customer service and profitability.

16. ORGANIZATIONS SHOULD FOCUS ON: Developing Accurate Forecasts: Strive for the most precise predictions possible. Collaborative Planning: Work closely with key supply chain partners to improve forecast accuracy. Information Sharing: Enhance supply chain visibility by providing real-time access to sales and inventory data. Rapid Communication: Quickly communicate about forecast errors, unplanned disruptions (like floods or work stoppages), and changes in plans to adjust operations promptly.

17. STEPS IN THE FORECASTING PROCESS Determine the purpose of the forecast . How will it be used and when will it be needed? This step will indicate the level of detail required in the forecast, the amount of resources (personnel, computer time, dollars) that can be justified, and the level of accuracy necessary. Establish a time horizon. The forecast must indicate a time interval, keeping in mind that accuracy decreases as the time horizon increases. Obtain, clean, and analyze appropriate data. Obtaining the data can involve significant effort. Once obtained, the data may need to be “cleaned” to get rid of outliers and obviously incorrect data before analysis. Select a forecasting technique.

18. 5. Make the forecast . 6. Monitor the forecast errors . The forecast errors should be monitored to determine if the forecast is performing in a satisfactory manner. If it is not, reexamine the method, assumptions, validity of data, and so on; modify as needed; and prepare a revised forecast.

19. FORECAST ACCURACY Accurate forecasting is crucial, but minimizing forecast errors is challenging due to the complex nature of real-world variables and the inherent randomness in data. Residual errors will always exist, so it's important to indicate how much a forecast might deviate from actual outcomes. This helps users understand the potential inaccuracy of forecasts. Decision-makers should consider forecast accuracy along with cost when selecting forecasting techniques. Reliable forecasts are essential for creating effective schedules and resource allocations. Inaccurate forecasts can lead to resource imbalances, production issues, increased costs, and customer dissatisfaction, making accurate forecasting vital for business success.

20. Two approaches to forecasting QUALITATIVE METHOD (FORECASTS WITH DATA) - Consists mainly of subjective inputs, which often defy precise numerical descriptions. QUANTITATIVE METHOD (WITH THE BASIS OF NUMBERS OF DATA) - involve either the projection of historical data or the development of associative models that attempt to utilize causal (explanatory) variables to make a forecast.

21. FORECASTING TECHNIQUES Judgmental forecasts rely on analysis of subjective inputs obtained from various sources, such as consumer surveys, the sales staff, managers and executives, and panels of experts. Quite frequently, these sources provide insights that are not otherwise available. Time-series forecasts simply attempt to project experience into the future. The techniques use historical data with the assumption that the future will be like the past. Some models merely attempt to smooth out random variations in historical data; others attempt to identify specific patterns in the data and project or extrapolate those patterns into the future, without trying to identify causes of the patterns. Associative models use equations that consist of one or more explanatory variables that can be used to predict demand. For example, demand for paint might be related to variables such as the price per gallon and the amount spent on advertising, as well as to specific characteristics of the paint (e.g., drying time, ease of cleanup).

22. FOUR QUALITATIVE TECHNIQUES EXECUTIVE OPINIONS Description: A small group of senior managers from different departments (e.g., marketing, finance) collaborates to develop a forecast. Example: A company’s top executives might forecast the demand for a new product based on their collective experience and market knowledge. Advantages: Leverages the expertise and diverse perspectives of senior managers. Disadvantages: Risk of one person's opinion dominating and less individual accountability. APPLICATION SAMPLE: Executive Opinions: During strategic planning, a retail chain’s executive team might forecast future market trends based on their collective experience and industry knowledge.

23. II. SALESFORCE OPTIONS: Description: Forecasts are based on input from sales staff or customer service representatives who have direct contact with customers. Example: Salespeople might estimate future sales based on their interactions with customers and recent sales trends. Advantages: Provides insights from those closest to the customers. Disadvantages: Sales staff may be influenced by recent experiences or conflicts of interest if forecasts impact their sales quotas. APPLICATION SAMPLE: Salesforce Opinions: A sales team at a software company might predict future software sales by assessing customer feedback and recent purchasing patterns.

24. III. CONSUMER SURVEYS: Description: Collecting opinions directly from consumers to gauge future demand or preferences. Example: A company conducts a survey to understand potential customer interest in a new product. Advantages: Gathers information directly from the source of demand. Disadvantages: Surveys can be expensive, time-consuming, and prone to biases or low response rates. APPLICATION SAMPLE: Consumer Surveys: A company launching a new beverage might survey potential customers to determine expected market acceptance and flavor preferences.

25. IV. Delphi Method: Description: An iterative process where a series of questionnaires is circulated among experts to reach a consensus forecast. Responses are anonymous to encourage honest feedback. Example: Experts might use the Delphi method to predict technological advancements, such as the widespread adoption of a new technology. Advantages: Encourages unbiased opinions and can handle complex, uncertain scenarios. Disadvantages: Can be time-consuming and relies heavily on the expertise and willingness of participants. APPLICATION SAMPLE: Delphi Method: Researchers might use the Delphi method to forecast when electric vehicles will become mainstream in the automotive industry, gathering input from industry experts and analysts.

26. FORECASTS BASED ON TIME- SERIES DATA Time-series forecasting involves analyzing data points collected at consistent intervals to predict future values. This approach assumes that future values can be estimated from past observations, without considering external variables.

27. KEY COMPONENTS OF TIME- SERIES DATA

28. 1. TREND Description: Long-term upward or downward movement in the data. Example: An increasing trend in monthly sales of electric vehicles over several years due to growing environmental awareness.

29. 2. Seasonality Description: Short-term, regular variations linked to specific time periods, such as days, weeks, or months. Example: Increased ice cream sales during summer months or higher retail sales during the holiday season.

30. 3. Cycles Description: Wavelike fluctuations lasting more than a year, often influenced by economic or political conditions. Example: Business cycle impacts on retail sales, where periods of economic growth or recession affect consumer spending.

31. 3. Irregular Variations Description: Unpredictable and unusual events, such as natural disasters or strikes, which can skew data. Example: A sudden drop in airline bookings due to a major airline strike.

32. 4. Random Variations Description: Residual variations remaining after accounting for trend, seasonality, and cycles, reflecting inherent randomness. Example: Minor fluctuations in daily traffic to a website that cannot be explained by trends or seasonal patterns.

33. Approaches to Time-Series Forecasting:

34. 1. Plotting and Visualization: Description: Graphing the time-series data to visually identify patterns like trends, seasonality, and cycles. Example: Plotting monthly sales data to detect an upward trend and seasonal peaks.

35. 2. Decomposition Description: Breaking down the time series into trend, seasonal, and irregular components to better understand underlying patterns. Example: Decomposing quarterly sales data into trend, seasonal effects (e.g., holiday sales spikes), and irregular events (e.g., one-time promotions).

36. 3. Moving Averages Description: Smoothing data by averaging over a specified number of periods to identify underlying trends and patterns. Example: Using a 12-month moving average to smooth out monthly sales data and highlight long-term trends.

37. 4. Exponential Smoothing Description: Applying weighted averages of past observations, with more recent observations given higher weights, to forecast future values. Example: Forecasting next month's sales using exponentially smoothed data from previous months, where recent months have more influence.

38. 5. Autoregressive Integrated Moving Average (ARIMA): Description: A statistical model that combines autoregressive, moving average, and differencing techniques to model time-series data. Example: Using ARIMA to forecast future monthly sales based on patterns observed in past sales data.

39. Important Considerations: Demand vs. Sales: Forecasting should focus on demand rather than sales figures, as sales may not fully reflect demand if stockouts or other issues occur. Handling Irregular Variations: Identify and adjust for irregular events to prevent distortion of the forecast.

40. Naive Forecasting Methods

41. Naive Forecasting is a straightforward approach where predictions are based on the most recent data point or a specific past value. This method is used for stable time series, those with seasonal variations, or those exhibiting trends

42. 1. Stable Series Forecasting Description: The forecast for the next period is simply the value of the most recent period. Example: If a product’s weekly demand was 50 cases last week, the forecast for this week would also be 50 cases.

43. 2. Seasonal Forecasting Description: The forecast for the current season is based on the value from the same season in the previous year. Example: To forecast turkey demand for this Thanksgiving, use last year's Thanksgiving demand as the forecast.

44. 3. Trend Forecasting Description: For series with a trend, the forecast is calculated by adding or subtracting the change observed in the last two periods to the most recent value. Example: If the last two values in a time series were 50 and 53, indicating a trend increase of 3 units per period, the forecast for the next period would be 53 + 3 = 56.

45. Advantages and Limitations: Advantages: Simplicity: Very easy to compute and understand. Cost-effective: Requires minimal computational resources and data analysis. Limitations: Accuracy: Often less accurate than more sophisticated methods, as it does not account for more complex patterns or changes in trends. Responsiveness: May not adjust well to sudden changes or fluctuations in data. Naive forecasting provides a baseline for comparing other forecasting methods. Despite its simplicity, it can be effective in situations where data patterns are consistent and changes are minimal.

46. OTHER FORECASTING METHODS:

47. 1. Focus Forecasting Concept : Focus forecasting is a method where multiple forecasting techniques (like moving averages, weighted averages, and exponential smoothing) are applied to recent historical data. The technique with the best accuracy is chosen for future predictions. Example : A company uses focus forecasting by applying different methods to the last few months of sales data. If exponential smoothing shows the lowest error for those months, it will be used to forecast next month's sales.

48. 2. Diffusion Models Concept : These models forecast the adoption and spread of new products or services based on how previous innovations were adopted. They consider factors such as market potential and media attention. Example : To forecast the success of a new tech gadget, a company uses diffusion models to estimate how quickly it will be adopted by consumers, based on similar past products and their market trajectories.

49. 3. Trend Analysis Concept : This involves identifying a trend (linear or nonlinear) in historical data and using it to project future values. A linear trend equation is often used, expressed as Ft= a+btFt = a + btFt = a+bt , where aaa is the intercept and bbb is the slope. Example : If a cell phone company finds a linear increase in weekly sales, they use a trend equation like Ft=45+5tFt = 45 + 5tFt=45+5t to predict future sales. For instance, if t=10t = 10t=10, the forecasted sales would be 45+5(10)=9545 + 5(10) = 9545+5(10)=95 units.

50. 4. Trend-Adjusted Exponential Smoothing Concept : This method combines exponential smoothing with a trend component. It's used when a time series shows a linear trend. It adjusts the forecast to account for the trend. Example : If cell phone sales show a trend and exponential smoothing alone would lag behind, trend-adjusted exponential smoothing combines smoothed errors and trend estimates to provide more accurate forecasts for future periods.

51. 5. Seasonality Concept : Seasonal variations are recurring changes in data linked to regular events (e.g., weather, holidays). Forecasting incorporates these patterns using seasonal indexes . Example : Retail sales of winter clothing increase in colder months. By using seasonal indices, a retailer can adjust forecasts to account for higher sales during winter. For instance, if sales are predicted to be 100 units, a seasonal index of 1.20 for winter would adjust the forecast to 120 units.

52. 6. Computing Seasonal Relatives Concept : Seasonal relatives express seasonal variations as percentages or quantities. They can be used to deseasonalize data or adjust forecasts for seasonality. Example : If the seasonal relative for July is 0.90, July sales forecasts would be adjusted to reflect 90% of the average monthly sales. For quarterly data, seasonal relatives might be 1.20, 1.10, 0.75, and 0.95 for the first through fourth quarters, respectively.

53. 7. CYCLES Concept : Cycles are long-term fluctuations that differ from seasonal patterns and are harder to predict. They often require identifying leading indicators or explanatory variables. Example : A company might use housing start data as a leading indicator to forecast demand for construction-related products. If housing starts to increase, it may predict higher future demand for products like appliances and furniture.

54. Monitoring Forecast Error

55. Monitoring forecast errors involves tracking and analyzing forecast errors to ensure forecasts are accurate and reliable. There are different methods to assess forecast errors, including control charts and tracking signals.

56. 1. Sources of Forecast Errors Model Inadequacy: The forecasting model might be missing key variables or unable to adapt to new trends or variables. Irregular Variations: Unexpected events like severe weather or shortages can impact forecast accuracy. Random Variations: Inherent randomness in data that cannot be predicted or controlled.

57. 2. Control Charts Purpose: To visually monitor forecast errors and determine if they are random or exhibit patterns. Construction: Errors are plotted over time with a center line (zero error) and control limits (typically ±2 standard deviations from the center). Example: If the Mean Squared Error (MSE) is 9.0, the standard deviation (s) is √9.0 = 3.0. The upper control limit (UCL) is 0 + 2(3.0) = +6.0, and the lower control limit (LCL) is 0 - 2(3.0) = -6.0.

58. 3. Tracking Signal Purpose: To detect bias in forecasts over time by comparing cumulative forecast errors to average absolute errors. Calculation: Tracking Signal = Cumulative Error / MAD (Mean Absolute Deviation). Example: If the cumulative forecast error is -11 and the MAD is 6.622, the tracking signal is -11 / 6.622 = -1.66, which is within the acceptable range of ±4.

59. 4. Practical Application: Control Chart Example: Track forecast errors over time and check if any errors fall outside the control limits or if patterns like trends or cycles are present. Tracking Signal Example: Compute MAD and cumulative errors to determine if the forecast shows a consistent bias.

60. USING FORECAST INFORMATION A manager can take a reactive or a proactive approach to a forecast. A reactive approach views forecasts as probable future demand, and a manager reacts to meet that demand (e.g., adjusts production rates, inventories, the workforce). Conversely, a proactive approach seeks to actively influence demand (e.g., by means of advertising, pricing, or product/service changes).

61. Computer Software in Forecasting Computers significantly enhance forecasting by enabling quick and efficient data analysis, reducing the need for manual calculations. Various software packages are available to support forecasting efforts, including: Excel Templates: These offer functionalities for methods like moving averages, exponential smoothing, linear trend equations, trend-adjusted exponential smoothing, and simple linear regression. For instance, an Excel template might automate the process of calculating a moving average, saving time and improving accuracy.

62. Operations Strategy and Forecasting Importance of Accurate Forecasts: Accurate forecasts are crucial for aligning supply with demand, enhancing profits, and improving customer service. Reliable short-term forecasts are especially valuable, as they build confidence in the forecasting process and allow organizations to focus more on strategic planning. Improving Forecast Accuracy: Shortening Time Horizons: Short-term forecasts are generally more accurate than long-term ones. By focusing on reducing lead times and building flexibility into operations, organizations can better respond to changing demands. Example: A company might streamline its supply chain to shorten delivery times for raw materials or adjust staffing levels more rapidly to align with demand fluctuations.

63. 3. Challenges with Forecasting: Long Lead Times: Forecasting becomes challenging for orders placed far in advance, especially when demand is influenced by unpredictable factors like weather. For example, garden supplies might be ordered months in advance based on anticipated spring weather, which is difficult to predict accurately. 4. Supply Chain Collaboration: Sharing Information: Enhancing forecast quality can be achieved by sharing demand data throughout the supply chain. For instance, companies like Hewlett-Packard and IBM require resellers to include forecast data in their contracts to improve accuracy and efficiency.