The Ten-Day MBA 4th Ed. Page 24

  With that information and the abbreviated table below, the average number of people waiting in line should be 8.1.

  A/MS: .45

  M = 1: .37

  M = 2: .23

  A/MS: .50

  M = 1: .50

  M = 2: .33

  A/MS: .60

  M = 1: .90

  M = 2: .67

  A/MS: .70

  M = 1: 1.60

  M = 2: 1.30

  A/MS: .80

  M = 1: 3.20

  M = 2: 2.90

  A/MS: .90

  M = 1: 8.10

  M = 2: 7.70

  It would seem logical that by adding a second express teller, the average line would be cut from 8 to 4, wouldn’t it?

  The waiting line would be reduced by over 97 percent by adding an extra teller. When the line is very busy, the second teller makes a big difference. Only queuing theory could tell you that. This teller problem is the simplest of examples. A whole “science” has been born around queuing. Academics have created books of tables and charts to answer many queuing dilemmas. Although you may not be an expert, you now know of the existence of queuing theory. That’s how most MBA courses work. They teach you the fundamentals, but they expect that as an MBA, you’ll seek out an expert to implement the program.

  INVENTORY

  The Balancing Act. The optimal inventory level is a delicate balancing act. Inventory decisions are tough because different departments of the same company have different goals. When it comes to automobiles, marketers prefer to have too much rather than too little inventory. Salespeople want product for their customers. They hate to lose a sale because they are out of the hot minivan or sports car. Finance people want to carry the least amount of inventory possible. A smaller inventory investment leaves them with more cash on hand for other investments or to pay higher dividends. Production departments like to run as efficiently as possible. Long runs reduce the waste of multiple starts and stops, but can, of course, also be responsible for significant inventory buildups. MBAs call the process for getting all this right supply chain management.

  Inventory Vocabulary. Inventory exists in one of three forms, be it in a factory or in a bakery:

  Raw Materials—Flour, sugar, shortening, ready-made icing, etc.

  Work in Process—Dough, pastry in the oven, pastry on cooling trays

  Finished Goods—Cakes, cookies, and doughnuts ready for sale

  Inventory includes not only the investment in materials, but also the investment in labor. As long as inventory remains in a company’s possession, money is being tied up. A simple and illustrative way of analyzing inventory levels is the inventory flow diagram below. It shows the type and value of a factory’s inventory. As a product is made, raw materials are combined with labor to create finished goods of higher value.

  INVENTORY FLOW DIAGRAM

  Reasons for Holding Inventory. There are five major and legitimate justifications for holding inventory:

  Pipeline—Inventory on hand to minimize production delays and maximize efficiency

  Cycle—Suppliers have minimum order amounts that are greater than immediate need.

  Safety—Stocks held to avoid a shortage because of uncertain production demands. Stockouts cost money when production is halted.

  Anticipatory—Inventory held in anticipation of known demand

  Speculative—Items purchased to beat supplier price increases

  In efficient companies, materials arrive just in time for production. This is called just-in-time inventory (JIT). The Japanese are famous for this. Factory line workers request parts as needed with inventory order cards called kanban. However, JIT does not necessarily mean that parts suppliers produce at the exact rate of the automaker’s assembly line needs. In reality the parts inventory sits in the warehouse of less powerful suppliers until it is called in by the auto manufacturers. True JIT has all manufacturing participants working in concert to meet production demands.

  Economic Order Quantity (EOQ). Special EOQ formulas help MBAs find just the right quantity of inventory to order to keep parts, raw materials, or shelf items to a minimum.

  The Economic Order Quantity formula is based on the trade-off of two costs associated with inventory.

  Carrying Costs—The costs associated with storage, insurance, and financing of inventory. The opportunity cost of using the company’s funds elsewhere should be considered.

  Ordering Costs—The costs of ordering that include all the accounting and clerical labor and materials associated with placing an order

  There are two extremes. A factory manager may choose to order huge quantities of parts infrequently, which reduces order costs, but maximizes carrying costs. Or he or she may order frequently to reduce carrying costs, maximizing ordering costs. The graph below shows that the least total cost is the inventory level when both ordering and carrying costs are minimized.

  THE BALANCE OF ORDERING AND HOLDING INVENTORY THE EOQ GRAPH

  The EOQ formula tries to find that optimal point at which the total cost of both ordering and carrying is minimized.

  The EOQ formula is:

  Where:

  Q* = Optimal inventory order quantity

  R = Annual unit requirements (Demand)

  O = Cost of placing an order

  C = Cost of carrying a unit of inventory per period

  Consider an auto parts distributor that supplies a Kansas City assembly plant with replacement lamp bulbs for car dome lights. Its sales history indicates that a level demand of 2,000 bulbs throughout the year is most likely. Each time the distributor orders a shipment from General Electric it costs $14 to process the order. A detailed study of costs reveals that it costs $.50 to carry each bulb in inventory for a year.

  The formula calculates the most economic inventory order as 335 bulbs. Since the demand is 2,000 bulbs, this means that there will be about six orders per year (2000/335). Sounds simple. But it is not. The simple EOQ formula only works if the demand is level. When demand fluctuates wildly throughout the year, as in the case of a grocery store’s demand for bagged ice, eggnog, or beer, the EOQ model has little value. Sophisticated computer programs perform a modified EOQ calculation more frequently to adjust the EOQ for fluctuating demand projections. In those situations the computer calculates varying optimal order sizes many times throughout the year. Even though the EOQ formula’s application is limited, an MBA can talk intelligently with inventory experts if a problem arises. Because when inventory piles up unexpectedly, it is serious business.

  “WE’RE STUCK WITH 700,000 BUSHELS OF CORN AND SOMEONE IS GOING TO HAVE TO EAT IT.”

  Material Requirements Planning (MRP): Inventory and Capacity Management. The knowledge of production scheduling and inventory control makes state-of-the-art manufacturing possible. MRP is a method for planning and controlling inventories required in a factory. Some say that MRP means “manufacturing resource planning,” but under any name, MRP is a sophisticated system to improve manufacturing efficiency. MRP schedules production and calculates the optimal amount of inventory needed for efficient production. With products that have many parts, such as automobiles, appliances, and electronics, such a calculation can only be arrived at by using a computer.

  To set up the system, the computer programmer must be familiar with the production process and material requirements. Then the computer can translate customer product demand into detailed orders to guide factory production and material requisitions from vendors.

  The MRP process begins when production engineers determine the most efficient production method. For autos like a Honda Civic, for example, the assembly line is the most efficient production method. The process investigation must include every step of assembly, from sanding the raw steel body to driving the Honda out of the factory. Time and motion studies, such as those Taylor conducted nearly a century ago, might be necessary. The capabilities of both machine and worker must be known to determine the capacity of the factory. For instance, production engineers
know exactly how many front quarter panels can be stamped out per hour and how many man-hours are required to operate the press.

  MATERIAL REQUIREMENT PLANNING SYSTEM

  Process engineers also have to detail all the part and material requirements of a product. The requirements list is called a Bill of Materials (BOM). It is recorded in the computer so that production demands can be “exploded” into exact material needs. For each Honda Civic, the MRP system would know that it needed two headlamps, forty-six two-inch screws, 4.2 quarts of paint, and hundreds of other parts. The inventories of the materials are also tracked by MRP. In that way MRP can direct the factory manager to keep adequate part inventories to feed the production line needs. At the same time, MRP minimizes inventory levels by telling inventory clerks to order economic order quantities.

  A “complete” MRP system, also called enterprise resource planning (ERP) software, coordinates the manufacturing process from forecasting customer demand, shipment of the finished product, managing the inventory in transit, and stocking the store shelf. The Master Production Schedule (MPS) within the computer sorts and stores all the information about demand, production, and materials and sends out orders to direct and coordinate manufacturing.

  STANDARDS AND CONTROL

  All the information about the production process necessary to create an MRP system or to use the other MBA efficiency tools provides the basis for the standards that managers use to measure and control performance. An MBA buzzword for using standards is the term benchmarking. This is where accountants jump in to help the operational side of the business. The managerial accounting section of the accounting chapter explained how accountants track and report production efficiency through the use of variances. The factory can vary by paying more than planned for materials (price variances) or using more materials or labor per unit produced (material and labor usage variance). By setting standards and seeing if they are met, production managers control the process.

  QUALITY

  Operations classes take the concept of standards a bit further and deal with the issue of “quality,” which is vital to America’s competitiveness. What is quality, anyway? Quality only means that the product or service “meets the standards” set by either the manufacturer or the consumer. Quality does not necessarily mean a flawless product or service. Nor does it mean the most expensive product in its class like a Rolls-Royce. Quality products perform “as expected.” Mundane things such as paper clips could be considered of high quality if they are not rusted and hold paper together well.

  There are three important “quality gurus” whose prescriptions are touted as the cure for America’s troubled manufacturing: Joseph Juran, W. Edwards Deming, and Philip Crosby. Each has made a fortune writing, lecturing, and consulting about quality.

  Juran and Fitness for Use. Joseph Juran uses the phrase “fitness for use” when speaking about quality. “Consumers should be able to count on the product for what they need or want to do with it.” Manufacturers should produce quality products while “achieving high yields and minimal downtime.”

  Fitness for use has five “dimensions”; quality of design, conformance to manufacturing standards, lack of breakdowns, satisfactory performance, and the ease of maintenance of product after purchase.

  Deming, TQM, and Statistical Process Control. W. Edwards Deming is famous for having taught the Japanese about quality in the 1950s, when American industry showed little interest in the subject. Deming quite simply proposed that quality could be achieved by identifying the causes of production problems throughout the process and by carefully monitoring production to stop errors before too many products were produced. Every step of the process is an opportunity for increased efficiency; hence the term Total Quality Management (TQM).

  He divided problems into two categories, “common causes” and “special causes.” Common causes are systemic problems, shared by many workers, machines, or product types. Special causes are those problems that relate to individual workers, machines, or material shipments.

  Deming, with the help of Juran and W.A. Shewhart, developed a tool for identifying problems called statistical process control (SPC). “It is unlikely that two parts, even when produced by the same operator on the same machine, would ever be identical. The issue, therefore, was distinguishing acceptable variations from variations that could indicate problems.” Statistical probability provides a method of making that distinction.

  Production engineers make that distinction by studying the expected tolerance of each production task. For example, the filling machine at a Coca-Cola bottling plant does not pour exactly two liters into the two-liter jugs. The range of error is a few milliliters above or below two liters. Production engineers need to perform detailed studies to determine the usual amount of liquid squirted into each bottle. This exercise will result in a determination of the bell curve or statistical frequency distribution of filling quantities. If you remember the normal or bell curve discussion in the QA chapter, the range of variation that occurs 68 percent of the time was called one standard deviation or one sigma from the expected quantity. Any production measure outside a one-sigma-tolerance quality standard would signal a production problem. If a production manager desires, he or she can choose two- or three-sigma tolerances. Six sigma also refers to a program coined by Motorola that refers to a goal of a six-sigma standard or 3.4 defects per million. Many companies have six-sigma programs in place to reduce defects and increase profits.

  In my Coca-Cola example, the production engineer selected a one-sigma tolerance and found that 68 percent of the time, the bottles measured in his or her sampling were filled within a range of ten milliliters above or below the desired two-liter level.

  Using Deming’s SPC, a filling-machine operator could take hourly batches of ten two-liter jugs off the assembly line. Using one-sigma tolerance, samples above two liters and ten milliliters would be above the upper control limit (UCL). For those measuring below two liters, 1,990 milliliters would be below the lower control limit (LCL). Measurements outside the limits would signal a “special problem,” meaning a feeding line is crimped or clogged. The process would be considered “out of control,” and the operator would be instructed to take corrective action. If after the correction the next samples are within the ten-milliliter tolerance, the process is “in control” and the machines are allowed to operate. (My Coca-Cola example of selecting the UCL and LCL was greatly simplified merely to expose you to the subject. The frequency and number taken in each sample by the operator greatly affects the statistical calculation of limits.) See chart, below.

  BELL CURVE OF COLA FILL QUANTITIES

  Using SPC, the filling-machine operator records his sample Coca-Cola measurements on SPC charts. On X Bar Control Charts the operator records the average (X Bar) of the sample measurement he or she takes every hour. The X Bar chart shows any tendency of the machine to drift or jump over time. If the chart is approaching a limit, the operator can investigate before the filling machine gets out of statistical control.

  The R (Range) Control Chart reveals any tendency of the process to behave more or less randomly over time. It measures the range between the largest and smallest measurement in the same sampling used to create the X Bar charts. Within each group sample, the average of the sample measurements might mask unacceptable deviations. For example, a sample of a one-liter measurement and a three-liter measurement would average to two liters and appear acceptable on an X Bar chart. However, it is safe to expect that customers would be upset with half-filled bottles as well as bottles sticky from being grossly overfilled. In the case of being outside of an R chart’s limit, the operator must also take corrective action.

  The hypothetical SPC X Bar and R charts of a twelve-hour bottling shift below highlight problems.

  STATISTICAL PROCESS CONTROL CHARTS COCA-COLA BOTTLING PLANT X BAR AND R CHARTS

  The sudden change in X suggests that there is a mechanical problem or a new employee unfamiliar with the specif
ications.

  The rise in R may signal that a machine may be deteriorating, a machine control may be vibrating and slipping out of specification, or a worker could be getting tired.

  Crosby and “Quality Is Free.” Philip Crosby’s claim to fame is the proclamation that “quality is free.” He believed that if manufacturers improved quality, “conforming to requirements,” total production costs would fall. Crosby proposed that the ultimate goal of a quality program is zero defects. Management must make a concerted effort to alter both the design and the production method to improve quality. In his opinion, any costs incurred in improving quality would be paid for by the saving of materials and labor that were once expended in correcting defects.

  Genichi Taguchi and Poor Quality Is a Crime. Japanese quality expert Genichi Taguchi was a key quality proponent in postwar Japan. He taught from a spiritual perspective that “making poor products is worse than a thief.” Society does not lose anything from a thief as it is a redistribution of wealth, but everyone loses when poor-quality products are made.

  HOT TOPICS

  With the basics of capacity, scheduling, standards, and control behind you, this chapter would not be complete without mentioning some of the trendy stuff that keeps popping up in the business press.

  CYCLE TIME

  The time it takes for a company to convert a product idea into a new product or to improve an already existing product is called the cycle time of introduction. In Detroit the design and retooling for a new automobile may take two or more years. The turnaround for a new fashion item is often six months from design to store delivery. The faster a company can turn out a new product to meet consumer demand, the better the corporation will compete in the marketplace. Accordingly, rapid cycle times are a competitive advantage and a hot MBA topic. Some trendy MBAs call the battle to act faster time-based competition.