The Art and Science of Forecasting in Operations Management

Making good estimates is the main purpose of forecasting. Every day, operations managers make decisions with uncertain outcomes. No one can see the future to know what sales will be, what will break, what new equipment will be needed, or what investments will yield. Yet those decisions need to be made and executed to move the firm forward.

What is Forecasting?

Forecasting is the art and science of predicting what will happen in the future. Sometimes that is determined by a mathematical method; sometimes it is based on the intuition of the operations manager. Most forecasts and end decisions are a combination of both.

Forecasting is conducted by what are referred to as time horizons.

1. Short range forecast. While it can be up to one year, this forecast is usually used for three months or less. It is used for planning purchases, hiring, job assignments, production levels, and the like.

2. Medium range forecast. This is generally three months to three years. Medium range forecasts are used for sales and production planning, budgeting, and analysis of different operating plans.

3. Long range forecast. Generally three years or more in time span, it is used for new products, capital expenditures, facility expansion, relocation, and research and development.

Medium and long range forecasts differ from short range forecasts by other characteristics as well.

1. Medium and long range forecasts are more comprehensive in nature. They support and guide management decisions in planning products, processes, and plants. A new plant can take seven or eight years from the time it is thought of, until it is ready to move into and become functional.

2. Short term forecasts use different methodologies than the others. Most short term forecasts are quantitative in nature and use existing data in mathematical formulas to anticipate immediate future needs and impacts.

3. Short term forecasts are more accurate than medium or long range forecasts. A lot can change in three months, a year, three years, and longer. Factors that could influence those forecasts change every day. Short term forecasts need to be updated regularly to maintain their effectiveness.

Types of Forecasts

There are three major types of forecasting, regardless of time horizon, that are used by organizations.

1. Economic forecasts address the business cycle. They predict housing starts, inflation rates, money supplies, and other indicators.

2. Technological forecasts monitor rates of technological progress. This keeps organizations abreast of trends and can result in exciting new products. New products may require new facilities and equipment, which must be planned for in the appropriate time frame.

3. Demand forecasts deal with the company's products and estimate consumer demand. These are also referred to as sales forecasts, which have multiple purposes. In addition to driving scheduling, production, and capacity, they are also inputs to financial, personnel, and marketing future plans.

Strategic Importance of Forecasts

Operations managers have two tools at their disposal by which to make decisions: actual data and forecasts. The importance of forecasting cannot be underestimated. Take a product forecast and the functions of human resources, capacity, and supply chain management.

The workforce is based on demand. This includes hiring, training, and lay-off of workers. If a large demand is suddenly thrust upon the organization, training declines and the quality of the product could suffer.

When the capacity cannot keep up to the demand, the result is undependable delivery, loss of customers, and maybe loss of market share. Yet, excess capacity can skyrocket costs.

Last minute shipping means high cost. Asking for parts last minute can raise the cost. Most profit margins are slim, which means either of those scenarios can wipe out a profit margin and have an organization operating at cost -- or at a loss.

These scenarios are why forecasting is important to an organization. Good operations managers learn how to forecast, to trust the numbers, and to trust their instincts to make the right decisions for their firm.

Forecasting System

These seven steps can generate forecasts.

1. Determine what the forecast is for.

2. Select the items for the forecast.

3. Select the time horizon.

Interested in learning more? Why not take an online Operations Management course?

4. Select the forecast model type.

5. Gather data to be input into the model.

6. Make the forecast.

7. Verify and implement the results.

Routinely repeat these steps, regardless of the time horizon, to stay abreast of changes in regard to internal and external factors.

Forecasting Approaches

There are two predominant approaches to forecasting: qualitative approach and quantitative analysis. A qualitative approach uses factors such as experience, instinct and emotion while the quantitative analysis relies heavily on mathematics, historical data and casual variables.

Qualitative methods include:

1. Jury of executive opinion. This is based on the inputs and decisions of high-level experts or management.

2. Delphi method. Decision makers, staff, and respondents all meet to develop the forecast. Every shareholder in the process provides input.

3. Sales force composite. Each sales person provides an individual estimate which is reviewed for realism by management, and then combined for a big picture view.

4. Consumer market survey. This is surveying the prospective customer base to determine demand for existing products and can also be used for new products.

As these methods are based mostly on instinct, experience and human input, be cautious of excessive optimism.

Quantitative methods are in two categories. Time-series models predict by assuming the future is a function of the past. Associative models uses similar historical data inputs and then includes other external variables such as advertising budget, housing, competitor's prices and more.

Time Series Models

Associative Model

Naïve method

Linear regression

Moving averages


Exponential smoothing


Trend projection


Service Sector

Service sector industries have other unique factors to incorporate into their forecasts. Local events can increase the need for hotel stays, food, gas, and more. Holidays will have an impact. It can be narrowed to hours in the day around popular meal times. Tools for forecasting in this regard include point of sale tracking that computes sales by the quarter hour to establish a pattern for scheduling of personnel for peak times and deliveries or other activities during slower periods.
Operations Scheduling

Operations scheduling focuses on jobs. Jobs are assigned to individuals for a period of time, or jobs are assigned to workstations for completion. A job is the objective being produced, either a good or a service. Scheduling to meet demand is a critical aspect of the operations manager's function in the organization.

"Scheduling is the process of organizing, choosing, and timing resource usage to carry out all the activities necessary to produce the desired outputs at the desired times, while satisfying a large number of time and relationship constraints among the activities and the resources." (Morton and Pentico) There are many ways to schedule and sequence jobs.

Performance Measures

Flow time is a performance measure that tracks the time a job is in the system. Past due is a measure of by how much time a job missed its due date. These are basic and very general measures to determine how to schedule a job and its priority. There are more refined techniques to aid in that determination:

1. Makespan is the total amount of time required to complete a group of jobs. It is calculated by subtracting the starting time of a job from the time of completion from the last job. This technique results in lower inventory and increased delivery speed.

2. Total inventory is the total when one adds the scheduled receipts for items, plus the on-hand inventories for those items, and reduces inventory holding costs.

3. Utilization is measured as a ratio of average output rate to maximum capacity. Maximizing utilization creates slack capacity.

These are all related somehow. For instance, by minimizing makespan, utilization is maximized. A combination of these techniques can be used to determine sequencing.


There are two types of environments in manufacturing: job shop and flow shop. The type of environment contributes to scheduling and sequencing decisions and methodology.

A flow shop uses continuous flow processes. These are most commonly found in medium- to high-volume production. All the jobs will have a similar flow pattern from workstation to work station. This shop benefits from the makespan technique. The group of jobs will be completed in the minimum amount of time, while maximizing utilization.

Johnson's rule is a dominant factor in flow shop scheduling. It is a procedure that demonstrates, with all workstations being equal in capability, all jobs should be given the same priority.

1. Scan workstation processing times and find the shortest processing time of the jobs awaiting processing.

2. Schedule the job to the workstation with the shortest processing time. If it's the first workstation, do it as early as possible. If it is a workstation further down the line, schedule it as late as possible.

3. Take out the just-scheduled job(s), and start the process over.

Job shop

A job shop is for low-to-medium volume and schedules its work by jobs or batches. They do not have linear flow to the work. Instead, requirements may vary the job routing. Since the unpredictability is so high, a job shop requires priority sequencing rules. The most common are First Come First Served (FCFS), or Earliest Due Date (EDD), to determine which jobs get the highest priority. In the event of a tie or other factors, other priority sequencing methods can be used to narrow it down. It may come down to just picking one job over another, if all else remains equal.

1. Critical ratio (CR) means the job with the lowest CR is completed next. The ratio is calculated by subtracting the due date from today's date, then dividing by how much shop time is left.

2. Shortest processing time means that the job that will take the shortest amount of time to complete is scheduled next.

3. Slack per remaining operations (S/RO). Slack means the amount of time left after considering processing time and due date. The job with the lowest S/RO is the next one up. It is calculated by subtracting today's date from the due date, and then to subtract the remaining shop time by that figure. Divide by number of jobs left to do to determine the S/RO.

Service Operations Scheduling

Service industries are different than manufacturing although they share a lot of the same principles. Scheduling is no different. Instead of job shop or flow shop, service functions are described as front office or back office.

Front office functions are divergent work flows like job shops. Demand fluctuates, is hard to predict, and requires scheduling to compensate for that. There is significant customer interaction and customization to complete those jobs.

Back office functions have lower customer interaction. Services are more standardized and a known quantity, much like a flow shop. Processes are similar to manufacturing processes -- repetitive and consistent, with little variation.

Labor Limitations

While workstations may be plentiful, workers to operate them may not. When the lacking resource is personnel, operations managers have to adjust their operations scheduling accordingly. Workers can be trained to operate more than one machine to generate some flexibility. It is a competitive edge to be able to change schedules quickly and keep everything moving smoothly along the supply chain.

1. Assign personnel to complete the job that has been in the system the longest.

2. Assign personnel to a workstation that has the most jobs waiting.

3. Assign personnel to jobs with the earliest due date.

4. Assign personnel to workstations with the most standard work to run.