2-A-2 Develop supply forecasts in light of economic, competitive, technology, market and currency trends and conditions that affect procurement
The purpose of forecasting is to develop supply forecasts in light of economic, competitive, technology, market and currency trends and conditions that affect supply management.
1) Purpose of forecasting the following:
A) Quantity - Businesses need to predict sales quantities for new and existing products and services so they can develop demand forecasts. While estimating expected sales dollars is common practice, demand must be predicted in terms of quantity and is needed for operational planning purposes. For example, organizations need to estimate material, labor, transportation, distribution and space requirements for the upcoming year. (Stanley and Matthews, 2008.) Price, availability, and the cyclical and seasonal characteristics are also considered when forecasting quantities of goods and services. The quantity demanded depends on the product's stage in its life cycle, as well as technological changes expected in the product and the process.
B) Industry capacity and availability - An organization must determine whether there will be enough industry capacity to meet overall customer demand and whether it will have the capacity to meet its share of the market. To create a capacity forecast, an organization evaluates the competition in terms of overlapping product lines and market coverage, and the expected effect on the marketplace. (Stanley and Matthews, 2008.) If there is a need to expand capacity with an international supplier, the supply management professional should be aware of the source of the increased capacity.
C) Cost or price - Supply management is often responsible for performing price forecasting. Price forecasting is most commonly used for commodities such as crude oil, metals and other raw materials. (Stanley and Matthews, 2008.) Situations for which this type of forecasting is important arise when the industries of the organization and its suppliers and customers are changing rapidly. When inputs to a finished product or service experience rapid change in price or availability, the costs to an organization and the prices charged to customers can be strongly affected. Prices can be affected by seasonality in demand patterns and by international conditions. Because of the volatility in many commodity markets, accurate forecasting of potential costs or the setting of a sale price can be complex and challenging.
D) Technology- Forecasting technology trends is typically done in an environment of uncertainty. Organizations need to forecast technology trends which can impact demand as products and services evolve. Technological advances can cut product life-cycle times. If an organization is not ready for those changes, a competitor most likely will be. Senior executives have to be aware of any technology gaps that may exist within their own organization and the potential threats to revenue or market share if their competitors adopt this technology.
E) Planning -The purchasing history and economic reviews of key spending categories should represent a rigorous and comprehensive analysis of key commodity categories, including a historical review of purchases, an analysis of the present competitive environment, forecasts of supply, demand and prices. All of these reviews should be the cornerstone of the planning activities for the supply management department and also can be used as a basic resource for other corporate planning activities. It is important to recognize that forecasting without the proper planning, research and integration of data will result in the organization being less effective than desired or required.
F) Assuring supply- Supply management has the responsibility to assure that supply can meet demand. In organizations that have made just-in-time GIT) supply arrangements, or when supply bases are reduced to one or a few suppliers for each significant product or service, such assurance becomes critical. Forecasting supply conditions in critical markets is important to the ability to continuously ensure supply of needed items.
2) Economic concepts and terms used in forecasting
A) Price indices - Price indices are found by a comparison of prices in one year to a base year. A price index is computed by expressing all items in the index in terms of relative change from a base period, and assigning a weight to each item in the given index. ("Forecast Basics," Purchasing, 1995,p.17.)
While there are many significant economic indices, those for producer prices, consumer: prices and the Gross Domestic Product, known as the GDP, or implicit price deflator, are among the best known in the United States. Most developed countries compute indices using economic data. There are many sources for worldwide economic data including The World Bank, the United Nations and economic research institutions at colleges and universities.
Custom indexing is a professional tool that organization use to measure, investigate and control price and cost changes within their own organization. Custom indices are especially useful for commodities that represent a large percentage of total purchases. Supply management professionals may create a supplier price index and use it to compare against the producer price index to determine if the supplier is asking for equitable price increases. The supply professional should be sure to identify any assumptions used to create the index.
1.0 Producer Price Index or PPI -The Producer Price Index (PPI), is a family of indices published by many national agencies, including the United States Bureau of Labor Statistics (BLS), The United Kingdom Office for National Statistics,2 and Statistics Finland3• The PPI, according to the BLS, "measures the average change over time in the prices received by domestic producers of goods and services." Note that the PPI is based on the selling price rather than the actual cost to produce an item. The PPI for commodities may also be adjusted for seasonality, weather, regular marketing and production cycles, model changes, seasonal discounts and holidays when there is some economic rationale for doing so and when statistical tests indicate that there is an effect present. These seasonally adjusted indices are used to analyze general price trends in the economy.
For manufacturing, these indices cover the different stages of production- raw materials, work-in-process and finished goods. Over 8,000 PPIs for individual and groups of products are published each month and they are the first measures of inflation released each month. As of February 2007, the Department of Labor added indices for the construction, tra9e, finance, transportation and service. industries, a practice used by Japan and being tested in the United Kingdom. Supply management professionals and their suppliers may use the PPI to predict price inflation and negotiate price escalation clauses. (Stanley and Matthews, 2008.) 2.0 Consumer Price Index or CPI -The Consumer Price Index, also published monthly by the U.S. department of Labor and other national agencies, is one of the most popular measures of price inflation for retail goods and services. The CPI is linked more closely to labor rates than the PPI. The CPI measures the average change in retail prices over time for eight major groups with more than 200 categories of various goods and services. The rate of price inflation is important because it affects how much consumers must pay for goods and services, the cost of doing business, and quality of life for those on fixed incomes. It helps businesses negotiate labor contracts, and governments to establish fiscal policy. However, it represents consumer purchases only, and does not reflect product or service substitutions that consumers might make. The CPI is not useful in predicting swings in the economy because it is a lagging indicator (explained below in section 2-A-2, 2C, 2.0).
3.0 Implicit Price Deflator- Implicit price deflator, also known as the GDP deflator, is an index used by the U.S. Department of Commerce. A deflator is a value that allows data to be measured over time in terms of some base period. An implicit price deflator is a broad measure of prices derived from separate estimates of real and nominal expenditures for the GDP or a subcategory of GDP. It is an index of prices for everything that a country produces, making it different from the CPI, which considers consumption only and includes prices of imports. The implicit price deflator was created by the U.S. Department of Commerce and compares the average level of prices in a given year to those of a base year. It is calculated by:
Implicit Price Deflator = Current-Dollar GDP (Nominal) / Real GDP
B) Interest rates- Interest is the term for the cost of borrowing money. An interest rate is the annual percentage of the principal that the borrower must pay for the use of borrowed funds. Interest rates vary depending on repayment risk, duration of loan and inflation expectations. Interest rates affect investment and stimulate or depress growth. High interest rates inhibit borrowing and investment or trigger price inflation, which allows business to pass on ~he high cost of money. Well known measures of short-, medium-, and long-term interest rates are the following: U.S. Treasury three-month bills, U.S. Treasury notes and bonds with maturities between three and 10 years, high-grade municipal bonds, corporate AAA bonds, Federal Reserve discount rate, prime rate charged by commercial banks and new-home mortgage rates. See the Federal Reserve Board at http:/ /www.federalreserve.gov /releases/ for current data on interest rates and money supply.
C) Economic indicators - Understanding economic indicators helps supply management professionals identify those market forces that will affect the supply and demand for a particular commodity, product or service and leads to n1ore effective material forecasts, supply management budgets and price negotiations. Some of these economic indicators are discussed in the following sections. No one indicator can give a true picture of the economy. The U.S. Bureau of Economic Analysis -(BEA) collects data on 120 indicators in three categories -leading, lagging and coincident - and develops each set into an index. An index is expressed in a way that indicates price changes over time. State and local governments develop their own indices for local business use.
BEA groups key indicators into composite indices of economic trends by their general tendency to turn or change direction before, during or after the general economy's cycles. This data is available at www.conference-board.org.
1.0 Leading -A leading indicator is a measure of economic activity that changes before the business cycle does and thus indicates its future direction. For example, bond yields often indicate the direction of the stock market. (ISM Glossary,2006.) It is any statistic that precedes changes in economic growth rates and business activity. Some examples include the change in the number of building permits issued in a given period, the money supply (the amount of cash and bank deposits held by organizations and households), inventory-level changes, average weekly hours of production workers, changes in stock prices and the number of unemployment insurance claims.
2.0 Lagging -A lagging indicator, on the other hand, is a measure of economic activity that tends to change after the state of the general economy has changed, for example the unemployment rate. (ISM Glossary, 2006.) A lagging indicator confirms there has been a change in the economy, and tends to follow changes in the economy. In other words, if the economy is improving, the lagging indicators will confirm that phenomenon, and vice versa. Some key lagging indicators include labor costs, business spending, prime interest rates, inventory book value, unemployment rates and outstanding bank loans.
3.0 Coincident -A coincident indicator is a measure of economic activity that changes concurrently with a change in the •business cycle. (ISM Glossary, 2006.) Some examples include personal income, non-agricultural employment and industrial production.
D) Inflation/ deflation - Inflation is an increase and deflation is a decrease in the economy's average level of prices. The rate of inflation is the percentage rate of increase in the average price level. High inflation signifies rapid price escalation and wage increases, while investment centers on tangible assets rather than on financial markets. As measured by the rate of change in the GDP deflator, the United States has not experienced deflation since the 1930s, but has experienced a persistent inflationary trend since the end of World War II.
E) Capacity utilization - Capacity utilization refers to the extent to which an organization, an industry or a nation uses its installed facilities. It compares actual output to potential output using the available installed capacity. The Capacity Utilization Index is computed each month in the U.S. by the Federal Reserve Board (FRB) in conjunction with the Industrial Production Index. The FRB calculates the proportion of plant and equipment used in the manufacturing, mining, and electric and gas utilities industries. This index represents the ratio of the Industrial Production Index to plant and equipment capacity. These figures are seasonally adjusted. This is an inferential statistic, derived from year-end surveys made by several governmental and private organizations. See the Federal Reserve Board at http:/ /www.federal reserve.gov /releases/ for current data on capacity utilization.
F) Economic indexing -An index is a comparison between a figure from a base period and a figure from a later period. This method summarizes and examines changes in the direction and level of economic activity. Index numbers relate such things as annual industrial production, industrial prices and consumer prices to those of base years.
An index typically starts with a base period or year defined. as 100. The base year is often selected as a period when the economy is relatively stable, without high inflation, recession or high unemployment. All other comparison years or periods are represented as percentage differences from that base year or period. For example, if 1999 was the base year (1 00) for a particular statistic, and the 2005 index number for that activity was 105, then the economic activity in 2005 would be 5 percent greater than the base period.
The formula for calculating percentage changes between two periods is:
(current period value -prior period value)/ prior period value*100%
For example, if2005 had an index number of 130, then comparing 2005 to 2001, the results would be: ((130-105)/105] X 100 = 23.8
Therefore, 2005 is 23.8 percent higher than 2001, in terms of 1999 (base year) units. G) Gross Domestic Product (GDP) – GDP is a measure of a nation’s domestic putput, which is the total value of all finished goods and services produced within a country within a given time period (usually one calendar year). It is a very general measure of economic activity. These are final goods and services, including only the value added at each process or distribution step. Consequently, they are purchases by households and governments, and not industry purchases. There are various con1ponents within GDP measuring activities (no in-process goods) that seek to adjust for depreciation, taxes and savings rates. Forecasting becomes more complex as organizations struggle to understand the nature of the economies of all of the countries in which they conduct business.
H) Balance of merchandise trade -The Merchandise Trade Balance is a measure of "visible" trade, which is trade in goods like cars and electronics. A positive value indicates a trade surplus (exports exceed imports).A negative value indicates a trade deficit (imports exceed exports). The phrase "balance of trade" refers to the difference between the value of a country's exports and the value of its imports. The balance for a given country is favorable if more merchandise is exported than imported.
I) Balance of payments -The balance of payments for a given country is a measure for the difference in the flows of funds in and out of a nation's boundaries. (ISM Glossary, 2006.) It is a summary statement of a country's transactions with the rest of the world during a year. It consists of the balance of trade plus the capital account, which measures the inflow and outflow of investments and loans, and the official reserve accounts that document the changes in government reserve accounts. The United States has been in an unfavorable trade and payments position for more than fifty years.
J) Exchange rates -What a buying organization will owe another organization based in another country is a function of the international exchange rates that exist at time of payment. Historically, exchange rates were determined under the gold standard. Each national currency was converted into a specified amount of gold. When the gold standard was abandoned, the trading nations adopted a system of fixed exchange rates. Today, international exchange rates "float" between parameters established by the central banks. Exchange rates respond to the market and to economic activity. While each country wants its currency to be "strong" relative to that of its trading partners, it does not want it to be too strong or its goods and services will be unattractive to countries whose currency is weaker.
For example, too strong of a U.S. dollar makes it more difficult to sell U.S. products abroad, because it takes more international currency to pay for U.S. products. In addition, too strong of a dollar reduces domestic investment because the dollar goes farther in international investment. Differential exchange rates are exchange rates imposed by a country's government that differ, depending upon the nature of the goods or services imported. A fixed exchange rate is a currency exchange rate set and maintained by a government. A variable, or floating, exchange rate is a currency exchange rate determined by market forces rather than government decisions. (ISM Glossary, 2006)
3) Sources of data used in forecasting
The generation of fact-based or quantitative forecasts begins with research and the accumulation of data. Many sources provide useful information, including governmental data and forecasts, private organization or public organization forecasts, commercial forecasts (made available to subscribers) and internal organization data and forecasts.
A) ISM Report On Business® - Manufacturing and Nonmanufacturing -The ISM Manufacturing and Non-Manufacturing Report On Business® comprise two surveys of approximately 350 organizations in 18 manufacturing industries and 18 non-manufacturing industries from around the country. Unlike other business system indicators, the ISM Report On Business® does not measure the level of activity, but the month-over-month change in activity level. The supply management professionals from these organizations are asked to report changes from the previous month for several activities. The main indicators for which change is reported are production, new orders, supplier performance, inventories, employment and commodity prices. Supply management professionals report whether each activity was higher/better than, the same as, or less/worse than the previous month. For each of these indicators, a "diffusion index" is formed by adding all of the "higher/better" responses with half of the "same" responses, to develop the indicator. Five of the manufacturing indexes are included in the ISM PMI composite index. A reading over 50 indicates the manufacturing economy is generally expanding. A reading below 50 indicates the manufacturing economy is generally contracting. Changes in the PMI index historically have had a close relationship to changes in GDP. Because the first government estimate of GDP for a quarter is not available until one month after the end of the quarter, the first month's PMI for the quarter gives an early indication of that quarter's GDP three months before the first government number is available.
A composite index for the non-manufacturing sector was first released in early 2008, and is comprised of four of the non-manufacturing indexes.
The ISM Report On Business® is among the most widely watched and highly regarded of all leading economic indicators. The report has two specific advantages. First, turns in diffusion (change) indices have the property of leading turns in the actual activities, when changes are measured by one quarter of a business cycle. Second, unlike governmental data, which are often several months old when released, the ISM data describe changes in the previous month's activity. For these reasons, the ISM Report On Business® is extremely valuable as a macro economic forecasting tool. More information on the manufacturing and non-manufacturing Report On Business® may be found at www.ism.ws.
B) Government publications (International and U.S. Domestic) - Federal, state, and local governmental agencies gather and publish a remarkable variety of information on a broad range of business and economic topics. In most metropolitan areas in the United States, there are libraries designated as federal depositories that contain information disseminated by the federal government. A document librarian is essential when researching government documents. Many universities, including those internationally, are a part of consortiums that share government publications from around the world that can be accessed by anyone. The U.S. Government Printing Office (GPO) is a source for many popular federal publications. The GPO Web site, http:/ /www.gpoaccess.gov /index.html, provides access to a wealth of electronic documents and links to other federal agency Web sites. Most countries have a wide array of national and international economic data available, however, the explanation of the data may be in the native language. Most embassies and consulates have sources of information since their mission is to enhance trade activities. In addition, the United Nations, World Trade Organization, International Monetary Fund and The World Bank, to name a few, could be useful in obtaining economic data. The U.S. Department of Commerce and state commerce departments offer many types of statistical information useful to forecasters. These include:
1.0 Survey of Current Business -The Survey of Current Business is published monthly by the Bureau of Economic Analysis of the U.S. Department of Commerce. This report provides estimates and analysis of U.S. economic activity. The survey also reviews current economic developments, and quarterly national income and product accounts. It is available at http:// wwv.bea.gov/scb/index.htm.
2.0 Federal Reserve Bulletin -The U.S. Federal Reserve Bulletin prepared by the Federal Reserve Board is a monthly publication that includes such information as flow of funds measures, interest rates, savings, wages, and prices. It is available at http:// www:federalreserve.gov I releases/ .
C) Private publications - Many associations, such as the Institute for Supply Management™ (ISM) and various industry associations, publish surveys or analyses of economic conditions for their members. Special-interest groups also generate information that would be useful. Other business forecasts can be found in The Wall Street Journal, Business Ukek or the Financial Times.
D) Commercial forecasts - Many organizations develop and market economic forecasts for sale to subscribers. Among these are banks and industry groups. Customized forecasts are available from many of these organizations.
E) Regional surveys - Many of the local affiliates of ISM publish their own business surveys. These surveys, which are similar in methodology to the ISM Report On Business®, reflect local or regional situations that may be significantly different from national conditions. An up-to-date listing of the regional ISM surveys is available at www.ism.ws. Each state in the United States also has economic data. The Europa World Yearbook and its Web site is a source for economic data on a variety 'of countries.
F) Internal historical data- Many organizations develop organization-specific forecasts from internal historical data. For example, forecasts of sales for individual products and families of products can be determined using internal records. Forecasts of standard costs, lead times, seasonality of demand, employment levels and financial data are additional types of internal historical information that can be gathered and used.
G) Industry sources - Industry publications, newsletters and associations specialize in particular industries. Examples include Chemical Ukek, American Metal Market, and The American Iron and Steel Institute.
H) Online indices and search engines/Internet - Most economic data can be found on the Internet at various industry, association and government Web sites. Much of this is available free of charge, but some require a subscription or membership. Some is in the form of a report that can be ordered. The Internet is an excellent source of international data that may be needed for forecasting supply and demand in countries around the globe, and understanding international conditions in the marketplace. International statistics are often segmented by developed nations, developing nations or newly industrialized countries, big emerging markets and poor third-world economies. Countries within a given classification are more likely to operate similarly and different from countries in another classification due to variations in income levels and capital available for international exchange. Three international organizations that provide extensive statistical data include the United Nations (UN) at http://www. un.org/, the Organization for Economic Co-Operation and Development (OECD) at http:/ /www.oecd.org/home/, and the International Monetary Fund (IMF) at http://www. imf.org/. The United Nations provides international trade statistics, and a world economics survey, among other publications. UN COMTRADE, for example, is a report on global commodity trade statistics for more than 150 countries. It is updated daily and covers more than 5,000 products with options for exploration, data mining, and visualization through graphics. It is available online at http://comtrade. un.org/. For more information see Stanley and Matthews, Effective Supply Management Performance, Chapter 4, ISM Professional Series, 2008.
4) Forecasting methodologies/techniques
In any comprehensive forecasting program, many methods of forecasting are applied. The technique chosen varies based on the type of forecast, the data available and relevant to the forecast, the time frame in which the forecast is needed, and the technology and budget available for the forecasting process. Several of the n1ost important characteristics and methods related to forecasting are discussed below:
A) Short-term vs. long-term forecasting- Long-term and short-term forecasts are often made. A short-term (usually up to one year) forecast is used as an aid in the development and execution of short-term purchase plans and operational or tactical activity. Long-term forecasts facilitate the evolution and development of strategic plans and typically include in-depth commodity and industry studies.
B) Macro vs.' micro forecasting -Macro forecasts project broad scale activities, such as a nation's gross national product or an industrial sector like the services sector or farm sector.
Micro forecasts are organization-specific, or limited to small segments of larger issues. Macro and micro forecasts usually are linked. An organization may forecast prices paid for a specific, locally-purchased product (micro), but that forecast will be dependent on the status of that supply industry nationally and globally (macro).An organization's projection for the next year's inventory investment is based on sales estimates, which, in turn, are based on general economic condition forecasts.
C) Delphi method - Many times, supply management professionals need forecasts of activities or issues for which no factual data exist. A procedure for developing such opinion-based forecasts is the Delphi method. First used by the Rand Corporation in 1963, this procedure begins with the identification of a panel of experts in the field of interest. Rather than bring these people together, they are deliberately kept apart and are unknown to each other. This is done because group dynamics and discussion may distort and reduce creativity. These experts are posed a series of questions regarding the topic. Each expert then prepares a written response to each question, along with supporting arguments. Each participant receives anonymous copies of all other responses. The experts are then invited to revise or defend their responses. The revisions are then submitted to the researcher, who repeats the process perhaps as many as three or four times, until a consensus develops. Uses for the Delphi method may includes sales revenues, including long-range forecasting for product and service demand, developing projections for new product demand, and predicting technological developments. The Delphi method may be used to forecast societal changes, scientific advances, the competitive environment or government regulation and may be useful in directing an organization's research and development. The results from use of the Delphi method can be used with other analytical tools to create a most-likely scenario forecast.
D) Correlation/ regression analysis - Correlation analysis and linear regression analysis are methods for measuring the statistical, but not causal, relationship between two variables. Measures of correlation are used to describe the degree of relationship between two variables (or data series). The correlation coefficient indicates whether the relationship is positive (both increase together) or negative (one increases while the other decreases). With the use of a regression equation, the value of one variable based on the other variable can be predicted. Correlation and regression analyses are particularly useful for forecasting estimates when historical d:ita exist and there is reason to believe that the current relationship between variables will continue.
E) Time-series analysis -Time-series analysis is a method for examining the factors that influence changes in data series over time. These factors include trends, cyclical variations, seasonal fluctuations and random influences. The underlying assumption is that a forecast can be created based on patterns observed within the time-series data. Time-series analysis for costs, prices, inventories, interest rates and employment is important to business because it can be used to extract information regarding variations that might be time-related. To accurately forecast based on time-series data, reasons for variation must be examined. To use time-series analysis to the best advantage, it must first be determined whether fluctuations in the observed data stem from seasonal variation, long-term general business or economic variation due to business cycles, long-term trends or are merely the result of random sampling . error. Sometimes all of these factors are causing variation in the data, and must be teased out of the data using statistical techniques. Data may be smoothed and seasonally adjusted to remove these influences, with the objective of exposing the long-term underlying relationship.
F) Central tendency - In any data set, the statistics of interest 1nay lie in the measures of central tendency, because these statistics allow generalized interpretation. Measures of central tendency include the mean, mode and median. The mean is the average of all of the values in the set. The mode is the most frequently appearing value in the set. The median is the midpoint when the set is arranged from lowest to highest value.
G) Variability - Central tendency shows only part of the picture. Measures of variability or dispersion, the degree to which values of the variables of interest differ from each other, are important. The smaller the measure of variability, the more tightly the numbers cluster around the central point. Measures of variability include range, variance and standard deviation. The range of a set of numbers is the largest number minus the smallest. The variance equals the squared sum of the differences of the individual values of a set from the mean, divided by the number of values in the set. The standard deviation is the square root of the variance. Both the measures of central tendency and the measures of variability are useful to describe a data set, and the relationships within a set of values for one variable. These measures may be used when a set of data contains values for more than one variable, to determine other statistics. One such statistic is the correlation coefficient between two variables which.is determined using correlation analysis described briefly in D above.
H) Analysis of cyclical data -A cyclical component of data is the residual variation fluctuating around the long-term trend, due to changes in the economic or business system. A cycle is longer than a season, often a multi-year phenomenon, and is tied to economic fluctuations, such as recessions or inflation. A cycle is among the most difficult patterns to predict. Economic indicators and indices are generally used to predict cycles. Seasonality is the name given to a pattern that is observed regularly, occurs for a constant length of time, and is influenced by given seasonal factors such as day of the week, or a given monthly quarter. Seasonality can be applied to hour-of-the-day variations, in which there are certain times of day that are always busier than others and can be used to forecast resources needed to handle peak demands in service industries. The existence of seasonality is usually easy to observe if there are enough data points to plot the data. For monthly data, a plot of two years of data (24 data points) will usually reveal seasonal patterns, if they exist. More data is better and can be used to confirm patterns. Since cycles can be much longer, sometimes as much as 50 years, it is a rare circumstance to have enough data to identify cycles by plotting the data.
Economic cycles and seasonal patterns can often make detection of long-term trends difficult. For this reason, data are often seasonally adjusted, meaning that the cyclical or seasonal component of the data is removed and the data is smoothed. The objective is to uncover the long range, underlying trend in the data to see if there is a sustainable increase or decrease in demand. (Trends are discussed in I below.)
Smoothing is a process that dampens variations in data to make trends more apparent. A moving average is applied to two or more periods of data, and a new set of data is created. For seasonal variations that appear to be quarterly, one may apply a three-month moving average to the data to smooth it. A statistical technique called exponential smoothing weights the observations being averaged to give more weight to the most recent data.
I) Trend analysis - An increasing or decreasing pattern of demand over time is considered a trend. A trend is the long-term component that underlies change in a series of data. Some of the factors that produce trends include population changes, productivity changes, technology changes, supply I demand changes and price changes. Trends may be. discerned by plotting smoothed data. The trend can be described statistically using exponential smoothing or regression analysis or other appropriate techniques. When trend and seasonality are present, the use of Winter's Model, {otherwise known as triple exponential smoothing), or seasonal exponential smoothing, is commonly applied. Another technique known as Box-Jenkins, is a heuristic whose goal is to find a mathematical formula that will approximately generate the historical demand patterns in a time series. This technique requires a great deal of data. For Winter's, Box-Jenkins, and other forecasting techniques, reliable statistical software is available. For more information on statistical models see Stanley and Matthews, Chapter 5, 2008.
J) Decision tree analysis -The decision tree is one means to help supply management professionals make interdependent decisions under uncertain conditions. A decision tree is a decision-making tool that maps alternative courses of action and their consequences. Its components include decision forks, outcome forks, outcome probabilities, outcome rewards and expected values. (ISM Glossary, 2006). Decision .trees are useful because they give some structure to decision-making and provide a more objective way of analyzing the alternatives. Computer software packages are available which make the process relatively easy.
5) Factors that can affect forecasts The numerical output of any forecasting tool is accurate only as long as conditions remain unchanged. Factors that can affect the accuracy of forecasts include war and threats of war, strikes and threats of strikes, and natural factors such as disasters, discoveries and depletions. Other issues for consideration include changes in technology, government, law and the population or consumer tastes. Each of these and other commodity-specific factors must be considered and tracked for each of an organization's critical raw materials or components.
A) Fluctuating lead times - Fluctuation in lead tin1es may result in material shortages or require larger inventories to protect the organization from delivery uncertainty. Both of these situations increase costs, so neither is acceptable in today's globally competitive economy. Factors causing lead time fluctuation range from financial and production problems, to demand increases, to constrictions in a supply chain. Many of these can be measured and forecasted. In planning material acquisitions, the supply management professional must carefully assess the probability of occurrence and causes of lead time fluctuations and work with suppliers to decrease or eliminate those fluctuations where possible.
B) Changing labor conditions -Any organization that purchases materials or services from organizations or supply chains that contain organized labor need to be aware of upcoming labor agreement expirations, and the accompanying potential for supply disruption. Not only can labor negotiations affect a supplier, they may disturb output from an entire industry. Strikes, or threats of strikes, can result in dramatic market changes. Advance warning of such potential supply constrictions allows a supply management professional to take protective action. Additionally, labor issues in countries or regions where the organization has suppliers must be carefully monitored.
A surplus or shortage of specific labor skills may also inhibit or enhance the ability of an organization to supply its customers. For example, a lack of skilled machinists or electronic technicians may restrict output of products requiring those skills. Availability of specific requisite labor skills should be observed and forecast in the same way as availability of other necessary inputs.
Strikes, threats of strike and shortages of properly trained skilled labor are elements of potential change to historical projections that the supply management professional must track and forecast for all critical purchased materials and components.
C) Changes in money markets - Early theories of international trade assumed that a country's factors of production (land, labor, and capital) were fixed and immobile. In today's economy, funds are transferred worldwide around the clock. Money markets in financial centers all over the world facilitate currency exchange and lending. Central banks operate in concert to regulate exchange-rate limits and interest-rate differences. Because many businesses have suppliers and customers around the globe, tracking these fluctuations may become the purview of the supply management professional.
In the United States, the Federal Reserve Board (the U.S. central bank) has the delicate job of balancing the availability of money in the national economy. Making funds more readily available has the benefit of driving interest rates down, but risks growth in the rate of inflation. Restricting money supply counteracts inflation, but results in higher interest rates that limit or restrict business expansion and affect unemployment.
At the micro or organizational level, an increase in interest rates obviously affects an organization's borrowing rate and its , cost of doing business. It may cause a project to be "crowded out;' to become economically unattractive to the organization. In an economic sense, reductions in investment lead to reductions in income and employment.
D) Political factors- Changes in government often portend changes in the business climate. Even in the United States' relatively stable political system, a change in the party in power may bring about change in emphasis on anti-trust law enforcement, environmental focus, economic goals regarding money supply, inflation and acceptable levels of unemployment.
In other countries, political change, or even the threat of political change, may pose significant risk to a multinational organization and outsourced supply. Supplies or prices of raw materials or components purchased from politically unstable countries may change precipitously. Forecasts and long-term purchase plans should always include political risk assessment (the U.S. Department of State, international banks and other multinational organizations publish country risk assessments that are useful to supply management professionals).
Even if purchases are not made from an international supplier, an examination of the countries of origin of purchased materials is likely to reveal significant import dependence. For this reason, supply management professionals should examine all strategic or critical materials and services to remain aware of political risk to supply continuity.
E) Technological shifts - In our current environment of rapidly changing and developing technology, shifts from an established technology to a new one can significantly affect forecasts. For example, the rapid development and adoption of the Internet as a means of locating sources, communicating with suppliers, and conducting business transactions caught many businesses unprepared in the late 1990s and early 2000s. That shift in technology came so rapidly that many established organizations paid little attention to it at first, giving new organizations opportunities to enter the market and take market share. Those organizations that have lost market share as a result were well aware of the impact of technological shifts on forecasts. While such shifts are difficult to foresee, and their impact even more difficult to predict, forecasters must be vigilant for continuing rapid technological developments and include their possibility in forecasts as alternate scenarios, whenever appropriate.
F) Climatic conditions - In the longer term, weather can play a critical role when considering the amount of sunlight and moisture necessary for materials originating with agricultural products. In the shorter term, it can adversely affect the ability to extract minerals and transport goods. At its extreme, weather can idle suppliers' plants due to flooding or wind damage, as well as interrupting their supply of power, water and gas utilities. In recent years there have been examples of extreme disasters such as tsunamis, floods, earthquakes and hurricanes that have shut down entire regions of the world. The potential for supply disruption can be mitigated to some extent by disaster planning on the part of supply management professionals.
G) Changes in global trade- Shifts can affect the relative supply and demand for goods an~ services, but also of importance are the sources precipitating these shifts. Changes in customs duties, trade embargoes, and import quotas can change availability, but political instability may limit supply, especially when civil unrest occurs. Supply management professionals also need to be aware of government subsidies that artificially lower the price of goods and the impact of developing nations acquiring higher levels of technological capability.