Forecasting, Planning and Scheduling Sir Eng R. L. Nkumbwa™ www.nkumbwa.weebly.com © 2010 Nkumbwa™. All Rights Reserved. 1 Outline Forecasting Production Planning Production Scheduling Material Resource Planning (MRP) Enterprise Resource Planning (ERP) © 2010 Nkumbwa™. All Rights Reserved. 2 Forecasting © 2010 Nkumbwa™. All Rights Reserved. 3 Forecasting Approaches Time Series Forecasting Methods Monitoring and Control Other Sectors © 2010 Nkumbwa™. All Rights Reserved. 4 What is Forecasting? Educated Guessing Process of predicting a future event Underlying basis of all business decisions Production Inventory Personnel Facilities © 2010 Nkumbwa™. All Rights Reserved. 5 Types of Forecasts by Time Horizon Short-range forecast Up to 1 year; usually less than 3 months Job scheduling, worker assignments Medium-range forecast 3 months to 3 years Sales & production planning, budgeting Long-range forecast 3+ years New product planning, facility location © 2010 Nkumbwa™. All Rights Reserved. 6 Short-term vs. Longer-term Forecasting Medium/long range forecasts deal with more comprehensive issues and support management decisions regarding planning and products, plants and processes. Short-term forecasting usually employs different methodologies than longer-term forecasting Short-term forecasts tend to be more accurate than longer-term forecasts. © 2010 Nkumbwa™. All Rights Reserved. 7 Influence of Product Life Cycle Introduction, Growth, Maturity, Decline Stages of introduction and growth require longer forecasts than maturity and decline Forecasts useful in projecting staffing levels, inventory levels, and factory capacity as product passes through life cycle stages © 2010 Nkumbwa™. All Rights Reserved. 8 Strategy and Issues During a Product’s Life Company Strategy/Issues Introduction Growth Best period to increase market share Practical to change price or quality image R&D product engineering critical Strengthen niche Maturity Poor time to change image, price, or quality Competitive costs become critical Decline Cost control critical Defend market position Fax machines Drive-thru restaurants CD-ROM 3 1/2” Floppy disks Sales Color copiers Station wagons Internet OM Strategy/Issues HDTV Product design and development critical Forecasting critical Standardization Little product differentiation Product and process reliability Frequent product and process design changes Cost minimization Competitive product improvements and options Less rapid product changes more minor changes Short production runs Increase capacity High production costs Shift toward product focused Limited models Enhance distribution Attention to quality © 2010 Nkumbwa™. All Rights Reserved. Optimum capacity Increasing stability of process Long production runs Product improvement and cost cutting Over capacity in the industry Prune line to eliminate items not returning good margin Reduce capacity 9 Types of Forecasts Economic forecasts Address business cycle, e.g., inflation rate, money supply etc. Technological forecasts Predict rate of technological progress Predict acceptance of new product Demand forecasts Predict sales of existing product © 2010 Nkumbwa™. All Rights Reserved. 10 Seven Steps in Forecasting Determine the use of the forecast Select the items to be forecasted Determine the time horizon of the forecast Select the forecasting model(s) Gather the data Make the forecast Validate and implement results © 2010 Nkumbwa™. All Rights Reserved. 11 Product Demand Charted over 4 Years with Trend and Seasonality Demand for product or service Seasonal peaks Trend component Actual demand line Random variation Year 1 © 2010 Nkumbwa™. All Rights Reserved. Year 2 Average demand over four years Year 3 Year 4 12 Actual Demand, Moving Average, Weighted Moving Average 35 Sales Demand 30 25 Weighted moving average Actual sales 20 15 10 Moving average 5 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month © 2010 Nkumbwa™. All Rights Reserved. 13 Realities of Forecasting Forecasts are seldom perfect Most forecasting methods assume that there is some underlying stability in the system Both product family and aggregated product forecasts are more accurate than individual product forecasts © 2010 Nkumbwa™. All Rights Reserved. 14 Forecasting Approaches Qualitative Methods Used when situation is vague & little data exist New products New technology Involves intuition, experience e.g., forecasting sales on Internet © 2010 Nkumbwa™. All Rights Reserved. Quantitative Methods Used when situation is ‘stable’ & historical data exist Existing products Current technology Involves mathematical techniques e.g., forecasting sales of color televisions 15 Overview of Qualitative Methods Jury of executive opinion Delphi method Panel of experts, queried iteratively Sales force composite Pool opinions of high-level executives, sometimes augment by statistical models Estimates from individual salespersons are reviewed for reasonableness, then aggregated Consumer Market Survey Ask the customer © 2010 Nkumbwa™. All Rights Reserved. 16 Jury of Executive Opinion Involves small group of high-level managers Group estimates demand by working together Combines managerial experience with statistical models Relatively quick ‘Group-think’ disadvantage © 2010 Nkumbwa™. All Rights Reserved. 17 Sales Force Composite Each salesperson projects his or her sales Combined at district & national levels Sales reps know customers’ wants Tends to be overly optimistic © 2010 Nkumbwa™. All Rights Reserved. 18 Delphi Method Iterative group process 3 types of people Decision makers Staff Respondents Reduces ‘group-think’ Decision Makers Staff (What will (Sales?) (Sales will be 50!) sales be? survey) Respondents (Sales will be 45, 50, 55) © 2010 Nkumbwa™. All Rights Reserved. 19 Consumer Market Survey Ask customers about purchasing plans What consumers say, and what they actually do are often different Sometimes difficult to answer © 2010 Nkumbwa™. All Rights Reserved. 20 Quantitative Forecasting Methods (Non-Naive) Quantitative Forecasting Associative Models Time Series Models Moving Average Exponential Smoothing © 2010 Nkumbwa™. All Rights Reserved. Trend Projection Linear Regression 21 What is a Time Series? Set of evenly spaced numerical data Obtained by observing response variable at regular time periods Forecast based only on past values Assumes that factors influencing past and present will continue influence in future Example Year: Sales: © 2010 Nkumbwa™. All Rights Reserved. 1998 78.7 1999 63.5 2000 89.7 2001 93.2 2002 92.1 22 Time Series Components Trend Cyclical Seasonal Random © 2010 Nkumbwa™. All Rights Reserved. 23 Trend Component Persistent, overall upward or downward pattern Due to population, technology etc. Several years duration © 2010 Nkumbwa™. All Rights Reserved. 24 Seasonal Component Regular pattern of up & down fluctuations Due to weather, customs etc. Occurs within 1 year © 2010 Nkumbwa™. All Rights Reserved. 25 Cyclical Component Repeating up & down movements Due to interactions of factors influencing economy Usually 2-10 years duration © 2010 Nkumbwa™. All Rights Reserved. 26 Random Component Erratic, unsystematic, ‘residual’ fluctuations Due to random variation or unforeseen events Union strike Tornado Short duration & nonrepeating © 2010 Nkumbwa™. All Rights Reserved. 27 Naive Approach Assumes demand in next period is the same as demand in most recent period e.g., If May sales were 48, then June sales will be 48 Sometimes cost effective & efficient © 2010 Nkumbwa™. All Rights Reserved. 28 Moving Average Method MA is a series of arithmetic means Used if little or no trend Used often for smoothing Provides overall impression of data over time Equation Demand in Previous n Periods MA n © 2010 Nkumbwa™. All Rights Reserved. 29 Moving Average Solution Time Response Yi 1998 1999 2000 2001 2002 2003 © 2010 Nkumbwa™. All Rights Reserved. 4 6 5 3 7 NA Moving Total (n=3) NA NA NA 4+6+5=15 6+5+3=14 5+3+7=15 Moving Average (n=3) NA NA NA 15/3=5.0 14/3=4.7 15/3=5.0 30 Weighted Moving Average Method Used when trend is present Older data usually less important Weights based on intuition Often lay between 0 & 1, & sum to 1.0 Equation WMA = Σ(Weight for period n) (Demand in period n) © 2010 Nkumbwa™. All Rights Reserved. ΣWeights 31 Disadvantages of Moving Average Methods Increasing n makes forecast less sensitive to changes Do not forecast trend well Require much historical data © 2010 Nkumbwa™. All Rights Reserved. 32 Exponential Smoothing Method Form of weighted moving average Weights decline exponentially Most recent data weighted most Requires smoothing constant () Ranges from 0 to 1 Subjectively chosen Involves little record keeping of past data © 2010 Nkumbwa™. All Rights Reserved. 33 Exponential Smoothing Equations Ft = At - 1 + (1-)At - 2 + (1- )2·At - 3 + (1- )3At - 4 + ... + (1- )t-1·A0 Ft = Forecast value At = Actual value = Smoothing constant © 2010 Nkumbwa™. All Rights Reserved. 34 Forecast Effects of Smoothing Constant Ft = At - 1 + (1- ) At - 2 + (1- )2At - 3 + ... Weights = = 0.10 = 0.90 © 2010 Nkumbwa™. All Rights Reserved. Prior Period 2 periods ago 3 periods ago (1 - ) (1 - )2 10% 9% 8.1% 90% 9% 0.9% 35 Choosing Seek to minimize the Mean Absolute Deviation (MAD) If: Then: Forecast error = demand - forecast MAD © 2010 Nkumbwa™. All Rights Reserved. forecast errors n 36 Regression © 2010 Nkumbwa™. All Rights Reserved. 37 Least Squares Values of Dependent Variable Actual observation Deviation Deviation Deviation Deviation Deviation Deviation Deviation Point on regression line Yˆ a bx Time © 2010 Nkumbwa™. All Rights Reserved. 38 Actual and the Least Squares Line © 2010 Nkumbwa™. All Rights Reserved. 39 Linear Trend Projection Used for forecasting linear trend line Assumes relationship between response variable, Y, and time, X, is a linear function Estimated by least squares method Minimizes sum of squared errors Yi a bX i © 2010 Nkumbwa™. All Rights Reserved. 40 Least Squares Equations Equation: ˆ i a bx i Y n Slope: x i y i nx y b i n x i nx i Y-Intercept: © 2010 Nkumbwa™. All Rights Reserved. a y bx 42 Using a Trend Line Year 1997 1998 1999 2000 2001 2002 2003 Demand 74 79 80 90 105 142 122 © 2010 Nkumbwa™. All Rights Reserved. The demand for electrical power at N.Y.Edison over the years 1997 – 2003 is given at the left. Find the overall trend. 43 San Diego Hospital – Inpatient Days 10200 1.06 Combined Forecast 10000 9800 1.04 Trend 1.02 9600 1 9400 0.98 Seasonal Index 9200 0.96 9000 0.94 8800 0.92 Jan Feb Mar © 2010 Nkumbwa™. All Rights Reserved. Apr May Jun Jul Aug Sep Oct Nov Dec 44 Linear Regression Model Shows linear relationship between dependent & explanatory variables Example: Sales & advertising (not time) Y-intercept Slope Y^ i = a + b X i Dependent (response) variable © 2010 Nkumbwa™. All Rights Reserved. Independent (explanatory) variable 45 Linear Regression Equations Equation: Yˆ i a bx i n Slope: b x i y i nx y i 1 n x i2 nx 2 i 1 Y-Intercept: © 2010 Nkumbwa™. All Rights Reserved. a y bx 47 Interpretation of Coefficients Slope (b) Estimated Y changes by b for each 1 unit increase in X If b = 2, then sales (Y) is expected to increase by 2 for each 1 unit increase in advertising (X) Y-intercept (a) Average value of Y when X = 0 If a = 4, then average sales (Y) is expected to be 4 when advertising (X) is 0 © 2010 Nkumbwa™. All Rights Reserved. 48 Random Error Variation Variation of actual Y from predicted Y Measured by standard error of estimate Sample standard deviation of errors Denoted SY,X Affects several factors Parameter significance Prediction accuracy © 2010 Nkumbwa™. All Rights Reserved. 49 Least Squares Assumptions Relationship is assumed to be linear. Plot the data first - if curve appears to be present, use curvilinear analysis. Relationship is assumed to hold only within or slightly outside data range. Do not attempt to predict time periods far beyond the range of the data base. Deviations around least squares line are assumed to be random. © 2010 Nkumbwa™. All Rights Reserved. 50 Standard Error of the Estimate n 2 y y i c S y,x i 1 n2 © 2010 Nkumbwa™. All Rights Reserved. n n n i 1 i 1 i 1 2 y i a y i b xi y i n2 51 Correlation Answers: ‘how strong is the linear relationship between the variables?’ Coefficient of correlation Sample correlation coefficient denoted r Values range from -1 to +1 Measures degree of association Used mainly for understanding © 2010 Nkumbwa™. All Rights Reserved. 52 Sample Coefficient of Correlation r n n n i i i n x i yi x i yi n n n n n x i x i n yi yi i i i i © 2010 Nkumbwa™. All Rights Reserved. 53 Coefficient of Correlation and Regression Model Y r=1 Y Y^i = a + b X i r = -1 Y^i = a + b X i X Y r = .89 Y^i = a + b X i X Y X r=0 Y^i = a + b X i X r2 = square of correlation coefficient (r), is the percent of the variation in y that is explained by the regression equation © 2010 Nkumbwa™. All Rights Reserved. 55 Guidelines for Selecting Forecasting Model You want to achieve: No pattern or direction in forecast error Error = (Yi - Yi) = (Actual - Forecast) Seen in plots of errors over time ^ Smallest forecast error Mean square error (MSE) Mean absolute deviation (MAD) © 2010 Nkumbwa™. All Rights Reserved. 56 Forecast Error Equations Mean Square Error (MSE) n 2 ˆ (y y ) i i forecast Mean MSEAbsolute i 1 Deviation (MAD) Mean Absolute nPercent Error (MAPE) n MAD | y i 1 i n n MAPE 100 © 2010 Nkumbwa™. All Rights Reserved. i 1 yˆ i | errors 2 n | forecast errors | n actual i forecast i actual i n 57 Tracking Signal Measures how well the forecast is predicting actual values Ratio of running sum of forecast errors (RSFE) to mean absolute deviation (MAD) Good tracking signal has low values Should be within upper and lower control limits © 2010 Nkumbwa™. All Rights Reserved. 58 Tracking Signal Equation RSFE TS MAD n y i yˆ i i MAD © 2010 Nkumbwa™. All Rights Reserved. forecast error MAD 59 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 2 100 95 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. -10 TS -10 RSFE = Errors = NA + (-10) = -10 60 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 2 100 95 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. -10 -10 TS 10 Abs Error = |Error| = |-10| = 10 61 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 2 100 95 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. -10 -10 10 TS 10 Cum |Error| = |Errors| = NA + 10 = 10 62 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 2 100 95 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. -10 -10 10 TS 10 10.0 MAD = |Errors|/n = 10/1 = 10 63 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 2 100 95 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. -10 -10 10 10 10.0 TS -1 TS = RSFE/MAD = -10/10 = -1 64 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 -10 2 100 95 -5 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. -10 10 10 10.0 TS -1 Error = Actual - Forecast = 95 - 100 = -5 65 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 -10 -10 2 100 95 -5 -15 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. 10 10 10.0 TS -1 RSFE = Errors = (-10) + (-5) = -15 66 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 -10 -10 10 2 100 95 -5 -15 5 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. 10 10.0 TS -1 Abs Error = |Error| = |-5| = 5 67 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 -10 -10 10 2 100 95 -5 -15 5 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. 10 10.0 TS -1 15 Cum Error = |Errors| = 10 + 5 = 15 68 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 -10 -10 10 2 100 95 -5 -15 5 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. 10 10.0 15 TS -1 7.5 MAD = |Errors|/n = 15/2 = 7.5 69 Tracking Signal Computation Mo Forc Act Error RSFE Abs Cum MAD Error |Error| 1 100 90 -10 -10 10 2 100 95 -5 -15 5 3 100 115 4 100 100 5 100 125 6 100 140 © 2010 Nkumbwa™. All Rights Reserved. TS 10 10.0 -1 15 -2 7.5 TS = RSFE/MAD = -15/7.5 = -2 70 Plot of a Tracking Signal Signal exceeded limit + Upper control limit 0 - Tracking signal Acceptable range Lower control limit Time © 2010 Nkumbwa™. All Rights Reserved. 71 160 140 120 100 80 60 40 20 0 3 2 Forecast 1 Actual demand 0 Tracking Signal -1 -2 Tracking Singal Actual Demand Tracking Signals -3 0 1 2 3 4 5 6 7 Time © 2010 Nkumbwa™. All Rights Reserved. 72 Forecasting in the Service Sector Presents unusual challenges special need for short term records needs differ greatly as function of industry and product issues of holidays and calendar unusual events © 2010 Nkumbwa™. All Rights Reserved. 73 Forecast of Sales by Hour for Fast Food Restaurant 20 15 10 5 0 +11-12 +1-2 11-12 12-1 1-2 2-3 © 2010 Nkumbwa™. All Rights Reserved. +3-4 +5-6 3-4 4-5 5-6 +7-8 +9-10 6-7 7-8 8-9 9-10 10-11 74 Planning © 2010 Nkumbwa™. All Rights Reserved. 75 Aggregate Planning Requires Logical overall unit for measuring sales and outputs Forecast of demand for intermediate planning period in these aggregate units Method for determining costs Model that combines forecasts and costs so that planning decisions can be made © 2010 Nkumbwa™. All Rights Reserved. 76 Planning Setting goals & objectives Example: Meet demand within the limits of available resources at the least cost Determining steps to achieve goals Example: Hire more workers Setting start & completion dates Example: Begin hiring in Jan.; finish, Mar. Assigning responsibility © 2010 Nkumbwa™. All Rights Reserved. 77 Planning Tasks and Responsibilities © 2010 Nkumbwa™. All Rights Reserved. 78 Relationships of the Aggregate Plan Marketplace and Demand Demand Forecasts, orders Product Decisions Process Planning & Capacity Decisions Aggregate Plan for Production Master Production Schedule, and MRP systems Research and Technology Work Force Raw Materials Available Inventory On Hand External Capacity Subcontractors Detailed Work Schedules © 2010 Nkumbwa™. All Rights Reserved. 79 What’s Needed for Aggregate Planning A mathematically based aggregate planning model requires considerable: time expertise problem definition model development model verification model application people who understand the problem people who understand both the modeling process, and the specific model money money to pay for all of the above often requires funding for several people for several months! © 2010 Nkumbwa™. All Rights Reserved. 80 Aggregate Planning Provides the quantity and timing of production for intermediate future Usually 3 to 18 months into future Combines (‘aggregates’) production Often expressed in common units Example: Hours, dollars, equivalents (e.g., FTE students) Involves capacity and demand variables © 2010 Nkumbwa™. All Rights Reserved. 81 Aggregate Planning Goals Meet demand Use capacity efficiently Meet inventory policy Minimize cost Labor Inventory Plant & equipment Subcontract © 2010 Nkumbwa™. All Rights Reserved. 82 Aggregate Planning Strategies Pure Strategies Capacity Options — change capacity: changing inventory levels varying work force size by hiring or layoffs varying production capacity through overtime or idle time subcontracting using part-time workers © 2010 Nkumbwa™. All Rights Reserved. 83 Aggregate Planning Strategies Pure Strategies Demand Options — change demand: influencing demand backordering during high demand periods counterseasonal product mixing © 2010 Nkumbwa™. All Rights Reserved. 84 Aggregate Scheduling Options Advantages and Disadvantages Option Advantage Changing inventory levels Changes in human resources are gradual, not abrupt production changes Varying workforce size by hiring or layoffs Avoids use of Hiring, layoff, other alternatives and training costs © 2010 Nkumbwa™. All Rights Reserved. Disadvantage Inventory holding costs; Shortages may result in lost sales Some Comments Applies mainly to production, not service, operations Used where size of labor pool is large 85 Advantages/Disadvantages - Continued Option Advantage Disadvantage Some Comments Varying production rates through overtime or idle time Matches seasonal fluctuations without hiring/training costs Permits flexibility and smoothing of the firm's output Overtime premiums, tired workers, may not meet demand Allows flexibility within the aggregate plan Loss of quality control; reduced profits; loss of future business Applies mainly in production settings Subcontracting © 2010 Nkumbwa™. All Rights Reserved. 86 Advantages/Disadvantages - Continued Option Advantage Disadvantage Some Comments Using part-time workers Less costly and more flexible than full-time workers Good for unskilled jobs in areas with large temporary labor pools Influencing demand Tries to use excess capacity. Discounts draw new customers. High turnover/training costs; quality suffers; scheduling difficult Uncertainty in demand. Hard to match demand to supply exactly. © 2010 Nkumbwa™. All Rights Reserved. Creates marketing ideas. Overbooking used in some businesses. 87 Advantage/Disadvantage - Continued Option Advantage Disadvantage Back ordering during highdemand periods May avoid Customer must overtime. Keeps be willing to capacity constant wait, but goodwill is lost. Counterseasonal Fully utilizes May require products and resources; allows skills or service mixing stable workforce. equipment outside a firm's areas of expertise. © 2010 Nkumbwa™. All Rights Reserved. Some Comments Many companies backorder. Risky finding products or services with opposite demand patterns. 88 The Extremes Level Strategy Chase Strategy Production rate is constant Production equals demand © 2010 Nkumbwa™. All Rights Reserved. 89 Aggregate Planning Strategies Mixed strategy Combines 2 or more aggregate scheduling options Level scheduling strategy Produce same amount every day Keep work force level constant Vary non-work force capacity or demand options Often results in lowest production costs © 2010 Nkumbwa™. All Rights Reserved. 90 Aggregate Planning Methods Graphical & charting techniques Popular & easy-to-understand Trial & error approach Mathematical approaches Transportation method Linear decision rule Management coefficients model Simulation © 2010 Nkumbwa™. All Rights Reserved. 91 The Graphical Approach to Aggregate Planning Forecast the demand for each period Determine the capacity for regular time, overtime, and subcontracting, for each period Determine the labor costs, hiring and firing costs, and inventory holding costs Consider company policies which may apply to the workers or to stock levels Develop alternative plans, and examine their total costs © 2010 Nkumbwa™. All Rights Reserved. 92 Forecast and Average Forecast Demand 22 © 2010 Nkumbwa™. All Rights Reserved. 18 21 21 22 20 93 Cumulative Demand Graph for Plan 1 Cumulative Demand (Units) 7,000 6,000 Reduction of inventory Cumulative level 5,000 production using 4,000 average monthly forecast requirements 3,000 Cumulative forecast requirements 2,000 1,000 Excess inventory Jan Feb Mar Apr May Jun © 2010 Nkumbwa™. All Rights Reserved. 94 Comparison of Three Major Aggregate Planning Methods Techniques Charting/graphical methods Transportation method Approaches Trial and error Simple to understand, easy to use. Many solutions; one chosen may not be optimal Optimization LP software available;permits sensitivity analysis and constraints. Linear function may not be realistic Heuristic Simple, easy to implement; tries to mimic manager’s decision process; uses regression Management coefficient model © 2010 Nkumbwa™. All Rights Reserved. Aspects 95 Controlling the Cost of Labor in Service Firms Seek: Close control of labor hours to ensure quick response to customer demand On-call labor resource that can be added or deleted to meet unexpected demand Flexibility of individual worker skills to permit reallocation of available labor Flexibility of individual worker in rate of output or hours of work to meet demand © 2010 Nkumbwa™. All Rights Reserved. 96 Making Yield Management Work Multiple pricing structures must be feasible and appear logical Forecast the use and duration of use. Manage the changes in demand. © 2010 Nkumbwa™. All Rights Reserved. 97 Hotel: Single Price Level Sales Demand Curve Potential customers exist who are willing to pay more than the $15 variable cost Passed up Some customers who profit paid $150 for the room contributions were actually willing to pay more $sales = Net price * 50 rooms =150*50 =$7500 Money left on the table $15 variable cost of room © 2010 Nkumbwa™. All Rights Reserved. $150 Price charged for room Price 98 Hotel: Two Price Levels Sales Net prices are: Price #1 => $85 Price #2 => $175 Demand Total sales = 1st net price *30 + 2nd net price *30 = $8100 $15 variable cost of room © 2010 Nkumbwa™. All Rights Reserved. $100 Price #1 $200 Price #2 99 Yield Management Matrix D u r a t I o n o f u s e Predictable use Unpredictable use © 2010 Nkumbwa™. All Rights Reserved. Fixed Price Variable Price Quadrant 1: Movies, stadiums, arenas, convention centers, hotel meeting space Quadrant 2: Hotels, Airlines, Rental cars,Cruise lines Quadrant 3: Restaurants,Golf courses, Internet service providers Quadrant 4: Continuing care hospitals 100 Scheduling © 2010 Nkumbwa™. All Rights Reserved. 101 Strategic Implications of Short-Term Scheduling By scheduling effectively, companies use assets more effectively and create greater capacity per dollar invested, which, in turn, lowers cost This added capacity and related flexibility provides faster delivery and therefore better customer service Good scheduling is a competitive advantage which contributes to dependable delivery © 2010 Nkumbwa™. All Rights Reserved. 102 Capacity Planning, Aggregate Scheduling, Master Schedule, and Short-Term Scheduling Capacity Planning 1. Facility size 2. Equipment procurement Aggregate Scheduling 1. Facility utilization 2. Personnel needs 3. Subcontracting Master Schedule 1. MRP 2. Disaggregation of master plan Short-term Scheduling 1. Work center loading 2. Job sequencing © 2010 Nkumbwa™. All Rights Reserved. Long-term Intermediate-term Intermediate-term Short-term 103 Short-Term Scheduling Examples Hospital Outpatient treatments Operating rooms University Instructors Classrooms Factory Production Purchases © 2010 Nkumbwa™. All Rights Reserved. 104 Forward and Backward Scheduling Forward scheduling: begins the schedule as soon as the requirements are known jobs performed to customer order schedule can be accomplished even if due date is missed often causes buildup of WIP Backward scheduling: begins with the due date of the final operation; schedules jobs in reverse order used in many manufacturing environments, catering, scheduling surgery © 2010 Nkumbwa™. All Rights Reserved. 105 Short-Term Scheduling Deals with timing of operations Short run focus: Hourly, daily, weekly Types Forward Scheduling B Backward Scheduling E Today © 2010 Nkumbwa™. All Rights Reserved. B Due Date Today E Due Date 106 The Goals of Short-Term Scheduling Minimize completion time Maximize utilization (make effective use of personnel and equipment) Minimize WIP inventory (keep inventory levels low) Minimize customer wait time © 2010 Nkumbwa™. All Rights Reserved. 107 Choosing a Scheduling Method Qualitative factors Number and variety of jobs Complexity of jobs Nature of operations Quantitative criteria Average completion time Utilization (% of time facility is used) WIP inventory (average # jobs in system) Customer waiting time (average lateness) © 2010 Nkumbwa™. All Rights Reserved. 108 Scheduling Methods Differ by Process ProcessFocused Variety of Methods © 2010 Nkumbwa™. All Rights Reserved. RepetitiveFocused ProductFocused Level Use Methods 109 Process-Focused Work Centers High variety, low volume systems Products made to order Products need different materials and processing Complex production planning and control Production planning aspects Shop loading Job sequencing © 2010 Nkumbwa™. All Rights Reserved. 110 Types of Planning Files Item master file - contains information about each component the firm produces or purchases Routing file - indicates each component’s flow through the shop Work-center master file - contains information about the work center such as capacity and efficiency © 2010 Nkumbwa™. All Rights Reserved. 111 Loading Jobs in Work Centers Assigning jobs to work centers Considerations Job priority (e.g., due date) Capacity Work center hours available Hours needed for job Approaches Gantt charts (load & scheduling) - capacity Assignment method - job to specific work center © 2010 Nkumbwa™. All Rights Reserved. 112 Gantt Load Chart Shows relative workload in facility Disadvantages Does not account for unexpected events Must be updated regularly Work Center M Metal Works Mechanical Electronics Painting Job 349 © 2010 Nkumbwa™. All Rights Reserved. T W F Job 350 Job G Job H Job D Job B Job C Th Job E Job I 114 Gantt Scheduling Chart Job Day 1S Day 2 Day Day Day Day T W 3 4 5T Day 6F Job A Job B Maintenance Scheduled activity time allowed Actual work progress Nonproduction time Job C Now © 2010 Nkumbwa™. All Rights Reserved. Day 7S Start of an activity End of an activity Point in time when chart is reviewed 115 Assignment Method Assigns tasks or jobs to resources Type of linear programming model Objective Minimize total cost, time etc. Constraints 1 job per resource (e.g., machine) 1 resource (e.g., machine) per job © 2010 Nkumbwa™. All Rights Reserved. 116 Assignment Method – Type Setter Example Typesetter Job A B C R-34 $11 $14 $6 S-66 $8 $10 $11 T-50 $9 $12 $7 Initial set-up © 2010 Nkumbwa™. All Rights Reserved. 117 Assignment Method - Four Steps Subtract the smallest number in each row from every number in that row; then subtract the smallest number in every column from every number in that column Draw the minimum number of vertical and horizontal straight lines necessary to cover all zeros in the table If the number of lines equals either the number of rows or the number of columns, then you can make an optimal assignment (Step 4) Otherwise: Subtract the smallest number not covered by a line from every other uncovered number. Add the same number to any number(s) lying at the intersection of any two lines. Return to Step 2 Optimal assignments will always be at the zero locations of the table © 2010 Nkumbwa™. All Rights Reserved. 118 Assignment Method – Type Setter Example Typesetter Job A B C R-34 $11 $14 $6 S-66 $8 $10 $11 T-50 $9 $12 $7 Initial set-up © 2010 Nkumbwa™. All Rights Reserved. 119 Step 1a & 1b Typesetter Job A B C Typesetter Job A B C R-34 5 8 0 R-34 5 6 0 S-66 0 2 3 S-66 0 0 3 T-50 2 5 0 T-50 2 3 0 Step 1a © 2010 Nkumbwa™. All Rights Reserved. Step 1b 120 Step 2 Typesetter Job A B C R-34 5 6 0 S-66 0 0 3 T-50 2 3 0 Smallest uncovered number © 2010 Nkumbwa™. All Rights Reserved. 121 Step 3 Typesetter Job A B C R-34 3 4 0 S-66 0 0 5 T-50 0 1 0 Make assignments © 2010 Nkumbwa™. All Rights Reserved. 122 Sequencing Challenge Order release Job Packet Job XYZ Production Control © 2010 Nkumbwa™. All Rights Reserved. Dispatch List Order Part Due Qty XYZ 6014 123 100 ABC 6020 124 50 Which job do I run next? Production 123 Sequencing Specifies order jobs will be worked Sequencing rules First come, first served (FCFS) Shortest processing time (SPT) Earliest due date (EDD) Longest processing time (LPT) Critical ratio (CR) Johnson’s rule © 2010 Nkumbwa™. All Rights Reserved. 124 Priority Rules for Dispatching Jobs FCFS First come, first served The first job to arrive at a work center is processed first EDD SPT LPT CR Earliest due date The job with the earliest due date is processed first Shortest processing time The job with the shortest processing time is processed first Longest processing time The job with the longest processing time is processed first Critical ratio The ratio of time remaining to required work time remaining is calculated, and jobs are scheduled in order of increasing ratio. © 2010 Nkumbwa™. All Rights Reserved. 125 First Come, First Served Rule Process first job to arrive at a work center first Average performance on most scheduling criteria Appears ‘fair’ & reasonable to customers Important for service organizations Example: Restaurants © 2010 Nkumbwa™. All Rights Reserved. 126 Shortest Processing Time Rule Process job with shortest processing time first. Usually best at minimizing job flow and minimizing the number of jobs in the system Major disadvantage is that long jobs may be continuously pushed back in the queue. © 2010 Nkumbwa™. All Rights Reserved. 127 Longest Processing Time Rule Process job with longest processing time first. Usually the least effective method of sequencing. © 2010 Nkumbwa™. All Rights Reserved. 128 Earliest Due Date Rule Process job with earliest due date first Widely used by many companies If due dates important If MRP used Due dates updated by each MRP run Performs poorly on many scheduling criteria © 2010 Nkumbwa™. All Rights Reserved. 129 Critical Ratio (CR) Ratio of time remaining to work time remaining Time remaining CR = Work days remaining = Due date - Today' s date Work (lead ) time remaining Process job with smallest CR first Performs well on average lateness © 2010 Nkumbwa™. All Rights Reserved. 130 Advantages of the Critical Ratio Scheduling Rule Use of the critical ratio can help to: determine the status of a specific job establish a relative priority among jobs on a common basis relate both stock and make-to-order jobs on a common basis adjust priorities and revise schedules automatically for changes in both demand and job progress dynamically track job progress and location © 2010 Nkumbwa™. All Rights Reserved. 131 Job Sequencing Example Job Job Work Processing time in days Job Due Date (day) A 6 8 B 2 6 C 8 18 D 3 15 E 9 23 © 2010 Nkumbwa™. All Rights Reserved. 132 Criteria to Evaluate Priority Rules ΣFlow times Average completion time # Jobs Process times Utilization Flow times Flow times Average number of jobs in the system Process times Av eragejob lateness © 2010 Nkumbwa™. All Rights Reserved. Late times Number of jobs 133 FCFS Parameter Average completion time Utilization Value Sequence 15.4 days A 36.4% B C Average number of jobs in the system 2.75 jobs Average job lateness 2.2 days © 2010 Nkumbwa™. All Rights Reserved. D E 134 SPT Parameter Average completion time Utilization Value Sequence 13 days B 43.1% D A Average number of jobs in the system 2.32 jobs Average job lateness 1.8 days © 2010 Nkumbwa™. All Rights Reserved. C E 135 EDD Parameter Average completion time Utilization Value Sequence 13.6 days B 41.2% A D Average number of jobs in the system 2.43 jobs Average job lateness 1.2 days © 2010 Nkumbwa™. All Rights Reserved. C E 136 LPT Parameter Average completion time Utilization Value Sequence 20.6 days E 27.2% C A Average number of jobs in the system 3.68 jobs Average job lateness 9.6 days © 2010 Nkumbwa™. All Rights Reserved. D B 137 Summary Rule Average Completion Time (days) Utilization (%) Average Number of Jobs in the System Average Lateness (Days) FCFS 15.4 36.4 2.75 2.2 SPT 13.0 43.1 2.32 1.8 EDD 13.6 41.2 2.43 1.2 LPT 20.6 27.2 3.68 9.6 © 2010 Nkumbwa™. All Rights Reserved. 138 Critical Ratio (CR) Job Job Work Processing time in days Job Due Date (day) Critical Ratio A 6 8 1.3 B 2 6 3.0 A C 8 18 2.25 D 3 15 5.0 E 9 23 2.56 © 2010 Nkumbwa™. All Rights Reserved. Sequence C E B D 139 Limitations of Rule-Based Dispatching Systems Scheduling is dynamic; therefore, rules need to be revised to adjust to changes in process, equipment, product mix, etc. Rules do not look upstream or downstream; idle resources and bottleneck resources in other departments may not be recognized Rules do not look beyond due dates © 2010 Nkumbwa™. All Rights Reserved. 140 Scheduling for Services Appointment systems - doctor’s office Reservations systems - restaurant, car rental First come, first served - deli Most critical first - hospital trauma room © 2010 Nkumbwa™. All Rights Reserved. 141 MRP & ERP © 2010 Nkumbwa™. All Rights Reserved. 142 Inventory Classifications Inventory Process stage Raw Material WIP Finished Goods © 2010 Nkumbwa™. All Rights Reserved. Number & Value Demand Type A Items B Items C Items Independent Dependent Other Maintenance Operating 143 Dependent versus Independent Demand Item Demand Source Material Type Method of Estimating Demand Planning Method Materials With Independent Demand Materials With Dependent Demand Company Customers Parent Items Finished Goods WIP & Raw Materials Forecast & Booked Customer Orders © 2010 Nkumbwa™. All Rights Reserved. EOQ & ROP Calculated MRP 144 Requirements for Effective Use of Dependent Demand Inventory Models Effective use of dependent demand inventory models requires that the operations manager know the: master production schedule specifications or bills-of-material inventory availability purchase orders outstanding lead times © 2010 Nkumbwa™. All Rights Reserved. 145 Inputs to the Production Plan Marketing Customer Demand Production Capacity Inventory Procurement Supplier Performance Aggregate Production Plan Management Return on Investment Capital © 2010 Nkumbwa™. All Rights Reserved. Finance Cash Flow Human Resources Manpower Planning Engineering Design Completion 146 The Planning Process Aggregate production plan Master production schedule Change requirements? Change capacity? Material requirements plan Change production plan? Change master production schedule? Detail capacity plan No Realistic Yes Is capacity plan being met? Execute capacity plans Is execution meeting the plan? Execute material plans © 2010 Nkumbwa™. All Rights Reserved. 147 Aggregate Production Plan Months January Aggregate Production Plan (shows the total quantity of amplifiers Weeks February 1,500 1 2 3 1,200 4 5 6 7 8 Master Production Schedule (Shows the specific type and quantity of amplifier to be produced 240 watt amplifier 150 watt amplifier 75 watt amplifier © 2010 Nkumbwa™. All Rights Reserved. 100 100 500 100 500 300 100 450 450 100 148 Typical Focus of the Master Production Schedule Make to Order Assemble to Order or Forecast Stock to Forecast (Process Focus) (Repetitive) (Product Focus) Schedule finished product Number of end items Typical focus of the master production schedule Schedule orders Number of inputs Motorcycles, autos, Print shop Machine shop TVs, fast-food Fine dining restaurant restaurant Examples: © 2010 Nkumbwa™. All Rights Reserved. Schedule modules Steel, Beer, Bread Light bulbs, Paper 149 Bill-of-Material List of components & quantities needed to make product Provides product structure (tree) Parents: Items above given level Children: Items below given level Shows low-level coding Lowest level in structure item occurs Top level is 0; next level is 1 etc. © 2010 Nkumbwa™. All Rights Reserved. 150 Product Structure for “Awesome” A © 2010 Nkumbwa™. All Rights Reserved. 151 Special Bills-of-Material Modular bills Modules are final components used to make assemble-to-stock end items Planning bills Used to assign artificial parent Reduces number of items scheduled Phantom bills Used for subassemblies that exist temporarily © 2010 Nkumbwa™. All Rights Reserved. 152 Bill-of-Material Product Structure Tree Bicycle(1) P/N 1000 Handle Bars (1) P/N 1001 Frame Assembly (1) P/N 1002 Wheels (2) P/N 1003 © 2010 Nkumbwa™. All Rights Reserved. Frame (1) P/N 1004 153 Time-Phased Product Structure Must have D and E completed here so production can begin on B Start production of D 1 week D 2 weeks to produce B 2 weeks E A 2 weeks 2 weeks E 1 week 1 week G 3 weeks F 1 week C D 1 2 © 2010 Nkumbwa™. All Rights Reserved. 3 4 5 6 7 8 154 Material Requirements Planning (MRP) Manufacturing computer information system Determines quantity & timing of dependent demand items © 2010 Nkumbwa™. All Rights Reserved. 155 MRP Requirements Computer system Mainly discrete products Accurate bill-of-material Accurate inventory status 99% inventory accuracy Stable lead times © 2010 Nkumbwa™. All Rights Reserved. 156 MRP Benefits Increased customer satisfaction due to meeting delivery schedules Faster response to market changes Improved labor & equipment utilization Better inventory planning & scheduling Reduced inventory levels without reduced customer service © 2010 Nkumbwa™. All Rights Reserved. 157 Structure of the MRP System BOM Master Production Schedule MRP by Period Report MRP by date report Lead Times Planned order report (Item Master File) Purchase advice Exception report Inventory Data Purchasing data © 2010 Nkumbwa™. All Rights Reserved. MRP planning programs (computer and software) Exception report 158 MRP and The Production Planning Process Forecast & Firm Orders Aggregate Production Planning Material Requirements Planning Master Production Scheduling Resource Availability No, modify CRP, MRP, or MPS Capacity Requirements Planning © 2010 Nkumbwa™. All Rights Reserved. Realistic? Yes Shop Floor Schedules 159 Master Production Schedule Shows items to be produced End item, customer order, module Derived from aggregate plan Example Item/Week Oct 3 Oct 10 Oct 17 Oct 24 Drills 300 200 310 300 Saws 300 450 310 330 © 2010 Nkumbwa™. All Rights Reserved. 161 Derivation of Master Schedule A S C B 10 11 15 8 B Lead time = 4 for A Master schedule for A Periods 5 6 40 7 Periods Gross requirements: B © 2010 Nkumbwa™. All Rights Reserved. 8 50 1 10 9 2 40+10 = 50 3 C Lead time = 6 for S Master schedule for S 4 5 9 10 11 12 13 40 20 30 6 40 50 20 7 8 15+30 = 45 Master schedule for S sold directly 1 2 10 10 3 Therefore, these are the gross requirements for B 162 MRP Dynamics Supports “replanning” Problem with system “nervousness” “Time fence” - allows a segment of the master schedule to be designated as “not to be rescheduled” “Pegging” - tracing upward in the bill-of-materials from the component to the parent item That a manager can react to changes, doesn’t mean he/she should © 2010 Nkumbwa™. All Rights Reserved. 163 MRP and JIT MRP - a planning and scheduling technique with fixed lead times JIT - a way to move material expeditiously Integrating the two: Small bucket approach and back flushing Balanced flow approach © 2010 Nkumbwa™. All Rights Reserved. 164 Lot-Sizing Techniques Lot-for-lot Economic Order Quantity Part Period Balancing Wagner-Whitin Algorithm © 2010 Nkumbwa™. All Rights Reserved. 165 Extensions of MRP Closed loop MRP Capacity planning - load reports MRP II - Material Resource Planning Enterprise Resource Planning © 2010 Nkumbwa™. All Rights Reserved. 166 Closed Loop MRP © 2010 Nkumbwa™. All Rights Reserved. 167 Extensions of MRP Capacity Planning Tactics for smoothing the load and minimizing the impact of changed lead time include: Overlapping - reduces the lead time, entails sending pieces to the second operation before the entire lot has completed the first operation Operations splitting - sends the lot to two different machines for the same operation Lot splitting - breaking up the order and running part of it ahead of the schedule © 2010 Nkumbwa™. All Rights Reserved. 168 Initial Resource Requirements/Smoothed Resource Requirements © 2010 Nkumbwa™. All Rights Reserved. 169 Extensions to MRP Enterprise Resource Planning MRP II with ties to customers and suppliers © 2010 Nkumbwa™. All Rights Reserved. 170 MRP and ERP Information Flows, Integrated with Other Systems © 2010 Nkumbwa™. All Rights Reserved. 171 MRP in Services Can be used when demand for service or service items is directly related to or derived from demand for other services restaurant - rolls required for each meal hospitals - implements for surgery etc. © 2010 Nkumbwa™. All Rights Reserved. 172 Product Structure, Bill of Materials, Bill of Labor for Veal Picante © 2010 Nkumbwa™. All Rights Reserved. 173 Distribution Resource Planning DRP requires: Gross requirements, which are the same as expected demand or sales forecasts Minimum levels of inventory to meet customer service levels Accurate lead times Definition of the distribution structure © 2010 Nkumbwa™. All Rights Reserved. 174