工作流參考模型英文

1、SECTION 1SCM TEMPLATE WORKFLOWSCM Template WorkflowRelease 4.2.1Copyright 2000 i2 Technologies, Inc.This notice is intended as a precaution against inadvertent publication and does not imply any waiver of confidentiality. Information in this document is subject to change without notice. No part of t

2、his document may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or information storage or retrieval systems, for any purpose without the express written permission of i2 Technologies, Inc.The software and/or database described i

3、n this document are furnished under a license agreement or nondisclosure agreement. It is against the law to copy the software on any medium except as specifically allowed in the license or nondisclosure agreement. If software or documentation is tobe used by the federal government, the following st

4、atement is applicable: In accordance with FAR 52.227-19 Commercial Computer Software Restricted Rights, the following applies: This software is Unpublishedrights reserved under the copyright laws of the United States.The text and drawings set forth in this document are the exclusive property of i2 T

5、echnologies, Inc. Unless otherwise noted, all names of companies, products, street addresses, and persons contained in the scenarios are designed solely to document the use of i2 Technologies, Inc. products.The brand names and product names used in this manual are the trademarks, registered trademar

6、ks, service marks or trade names of their respective owners. i2 Technologies, Inc. is not associatedwith any product or vendor mentioned in this publication unless otherwise noted.The following trademarks and service marks are the property of i2 Technologies, Inc.: EDGE OF INSTABILITY; i2 TECHNOLOGI

7、ES; ORB NETWORK; PLANET; and RESULTS DRIVEN METHODOLOGY. The following registered trademarks are the property of i2 Technologies, Inc.: GLOBAL SUPPLY CHAIN MANAGEMENT; i2; i2 TECHNOLOGIES and design; TRADEMATRIX; TRADEMATRIX and design; and RhythmLink.February, 2000Document ID: HiTech 4.2 SCM Templa

8、te WorkflowDocument Version:V 1.0Document Title:HiTech 4.2 SCM Template WorkflowDocument Revision:Draft 1Revision Date:3 February, 2000Document Reference:.Primary Author(s):SCM Team Krishnan Subramanian, Jatin Bindal, Abhay SinghalComments:ContentsSCM PROCESSES OVERVIEWSCM ProcessesDEMAND PLANNINGDe

9、mand ForecastingTop-Down ForecastingBottom-Up ForecastingLife Cycle Planning New Product Introductions and Phase-In/Phase-OutEvent PlanningConsensus ForecastAttach-Rate Forecasting/Dependent Demand Forecasting in Configure-to-Order environmentsDemand CollaborationFlex Limit PlanningForecast NettingF

10、orecast ExtractionMASTER PLANNINGSupply PlanningEnterprise Planning: Inventory PlanningEnterprise planning: Long term capacity planningEnterprise planning: Long term material planningFacility Planning: Supply plan for enterprise managed componentsCollaboration Planning for Enterprise and Factory Man

11、aged Components Procurement CollaborationCollaboration Planning with Transportation Providers - Transportation CollaborationAllocation PlanningDEMAND FULFILLMENTOrder PromisingPromising new ordersConfigure to Order (CTO) OrdersBuild to Order (BTO) OrdersOrder PlanningFactory PlanningTransportation P

12、lanningSCM Processes OverviewThe following figure briefly describes the solution architecture for the core processes that constitute the SCM solution. SCM ProcessesThe SCM template as a whole performs the following functions:1. Demand Planning: Forecasting and demand collaboration. Sales forecasts a

13、re generated using various statistical models and customer collaboration.2. Master Planning: Long term and medium term master planning for material as well as capacity. Master planning can be done at both the enterprise level (for critical shared components) and the factory level. In addition, decis

14、ions relating to material procurement and capacity outsourcingof materials from suppliers (or capacity outsourcing decisions) can be made.3. Allocation Planning: Reserving product supply for channel partners or customers based on pre-specified rules. Also, managing the supply so that orders that hav

15、e already been promised can be fulfilled in the best possible manner (on the promised dates and in the promised quantities).4. Order Promising: Promising a date and quantity to customer orders. These promises are made looking at the projected supply. In addition, sourcing decisions are also made her

16、e after considering such variables as lead-time, product cost, shipping cost, etc.5. Order Planning: Detailed order planning encompassing multiple factories. In addition detailed transportation planning is also done which can handle such complex requirements as merging two shipments from different l

17、ocations during transit.Information flows seamlessly between all these functions. The inputs to the system are the static data (supply chain structure, supplier relationships, seller and product hierarchies, supplier relationships, etc), some forecast data and actual orders. The output is a comprehe

18、nsive and intelligent supply chain plan which takes all the supply chain delivery processes into consideration in order to maximize customer satisfaction, at the same time reducing order fulfillment lead times and costs.The scope of this document is to describe the scenarios modeled as a part of the

19、 current release of the template (Hitech2). For any planning system, the place to begin planning is demand forecasting. We look at this in more detail in the next section.Demand PlanningThe objective of the Demand Planning process is to develop an accurate, reliable view of market demand, which is c

20、alled the demand plan. The Demand Planning process understands how products are organized and how they are sold. These structures are the foundation of the process and determine how forecast aggregation and disaggregation is conducted. A baseline statistical forecast is generated as a starting point

21、. It is improved with information directly from large customers and channel partners through collaboration. The forecast is refined with the planned event schedule, so the demand plan is synchronized with internal and external activities. Each product is evaluated based on its lifecycle, and continu

22、ally monitored to detect deviation. New product introductions are coordinated with older products, pipeline inventories, and component supply to maximize their effectiveness. Attach rates are used to determine component forecasts given the proliferation of products. The result is a demand plan that

23、significantly reduces forecast error and calculates demand variability, both of which are used to determine the size of the response buffers. The specific response buffers and their placement are different based on the manufacturing model employed, therefore the Demand Planning process must represen

24、t those differences.Order PlanningDemand PlanningOrder PromisingAllocationPlanningDemand ForecastingTop down forecastingBottom up forecastingLife cycle planningOption forecast Consensus forecastingForecast extractionDemand CollaborationDemandPlanningCustomersOrder Creation& CaptureForecast Netti

25、ngMaster PlanningThe following figure identifies the key processes that constitute demand planning and the scenarios that are modeled in the template.Demand ForecastingTop-Down ForecastingDefinitionTop down forecasting is the process of taking an aggregate enterprise revenue target and converting th

26、is revenue target into a revenue forecast by sales unit/product line. This allocation process of revenue targets can be done using historical performance measures or using rule based allocation techniques. The revenue targets can further be broken down into unit volume forecasts by using Average Sel

27、ling Price information for product lines.Historical information is typically more accurate at aggregate levels of customer/product hierarchies. Therefore, statistical forecasting techniques are typically applied at these aggregate levels. At levels where historical information might not be very rele

28、vant or is not perceived to be accurate, this allocation can be done with a rule-based approach. Frequency: This process is typically performed at a monthly/quarterly frequency, with the forecast being generated for the next several months/quarters. Scenario DescriptionBased upon historical bookings

29、 at an aggregate level across the entire company (for all products and geographys), the system will automatically generate multiple forecasts using different statistical techniques. The statistical techniques will account for such things as seasonality, trends, and quarterly spikes. Each statistical

30、 forecast will be compared with actuals to calculate a standard error. This will automatically occur at every branch (intersection) in the product and geographic hierarchies. The aggregate statistical forecast generated for the entire company will be automatically disaggregated at every intersection

31、 using the statistical technique with the smallest standard error. The outcome of this process will be a “Pickbest” statistically generated forecast at every level in the product and geography hierarchies. This forecast is then used as a baseline or starting point.Inputs · Historical Bookings b

32、y units· Historical Statistically based Bookings ForecastOutputs· Multiple Statistical forecasts· Statistical “Pickbest” forecast· Forecast committed to top-down forecast database row.Benefits· Easy disaggregation of data means faster, more accurate forecasting· Simple

33、alignment of revenue targets· Uses top down statistical advantages to easily tie lower level forecasts to revenue targetsi2 Products Used TRADEMATRIX Demand PlannerBottom-Up ForecastingDefinitionThis process enables the different sales organizations/sales reps/operations planners to enter the b

34、est estimate of the forecast for different products. This process consolidates the knowledge of sales representatives, local markets, and operational constraints into the forecasting process. This forecast can be aggregated from bottom up and compared to the targets established by the top-down forec

35、asting process at the enterprise level. This will enable easy comparison between sales forecasts and financial targets. Frequency: This is a weekly process. However, there is continuous refinement of the forecast at an interval determined by the forecasting cycle time and/or nature of the change req

36、uired.Scenario DescriptionIn parallel with the top-down forecast, the sales force/operational planners will enter forecasts for independent demand for a particular SKU or product series by customer or region as is pertinent to a particular Product / Geography combination. This data will automaticall

37、y be aggregated and compared to the targets established by the top-down forecasting process. Using the Average Selling Price for a unit, the unit based forecasts can be converted to revenue dollars and automatically aggregated.The bottom-up forecast can also be generated using collaborative demand p

38、lanning with a customer. In this case, the consensus forecast for a product/product series for a customer is aggregated and compared to the top-down target. Input o Sales force inputo Operations Planning Input o Average Selling Price (ASP)o Customer forecast (from the Demand Collaboration process)Ou

39、tputs o Aggregated Sales forecast by unito Aggregated Sales Forecast by Dollarso Aggregated Operations Plan by unitBenefitso Automatic aggregation of data means faster, more accurate forecastingo Simple alignment of lower level Sales plans to higher level revenue targetsi2 Products UsedTRADEMATRIX D

40、emand Planner, TRADEMATRIX Collaboration PlannerLife Cycle Planning New Product Introductions and Phase-In/Phase-OutDefinition Forecasting product transitions plays a critical role in the successful phasing out and launch of new products. New Product Introduction (NPI) and phase In/phase out forecas

41、ting allows the enterprise to forecast ramp downs and ramp ups more accurately. Ramping can be defined in terms of either a percentage or as units. Typically new products are difficult to forecast because no historical information for that product exists. NPI planning must allow for new product to i

42、nherit historical information from other product when it is expected that a new product will behave like the older product. In situations where a new product will not behave like any other older product, NPI planning allows a user to predict a life cycle curve for a product, and then overlay lifetim

43、e volume forecasts across that curve.Scenario Description Given a forecast for two complimentary products, the user can change the ramping percentage of both to reflect the ramping up of one product and the ramping down of another. Given a New Product Introduction that is predicted to behave like an

44、 older product, the user can utilize historical data from the older product to be used in predicting the forecast for the new product. The scenarios for this process are executed in TradeMatrix Demand Planner. Future releases of the template will use TradeMatrix Transitional Planner to do product li

45、fe cycle planning.Inputso Historical bookingso New product and association with the older parto Product ramping information for a new productOutputs· Adjusted Forecast ramping broken out by % · New product forecast based on a similar products history· New product forecast based on lif

46、e cycle inputBenefits · The ability to forecast a new product using history from an another product· The ability to forecast using product life cycle curves· Cleaner product transitions allowing for decreased inventory obsolescencei2 Products UsedTRADEMATRIX Demand Planner, TRADEMATRI

47、X Transition PlannerEvent PlanningDefinition This process determines the effect of future planned events on the forecast. The marketing forecast is adjusted based on events related factors. A promotional campaign or price change by the company or the competition is an example of an event related fac

48、tor that may influence demand. The marketing forecast is adjusted up or down by a certain factor. The factor can be increased or decreased across periods to simulate a ramp-up or a ramp-down in sales depending upon the nature of the event. Frequency: Event BasedScenario Description An event row will

49、 model the influence of the event that will change the marketing forecast. A promotional campaign or price change by the company or the competition is an example of a factor that may influence demand. The user will populate the Event row with scalar values which when multiplied by the Marketing stat

50、istical forecast will adjust the Marketing forecast up or down by a factor (0.90 for a 10% decline or 1.05 for a 5% increase etc.). Event row can be increased or decreased across periods to simulate a ramp-up or a ramp-down in sales depending upon the nature of the event. Inputso Event constant fact

51、or typicallyo Historical Bookingso Marketing forecastOutputso Adjusted Marketing Forecast Benefits o The ability to allow events to dynamically influence forecastI2 Products UsedTRADEMATRIX Demand PlannerConsensus ForecastDefinitionThe consensus process is one in which the multiple forecasting proce

52、sses thus far used are brought together to arrive at one single forecast. All information critical to reaching consensus on the forecast will be brought together for analysis and facilitation of the consensus process. The level at which the consensus process is performed is typically at an intermedi

53、ate level, where the forecast is most meaningful for the different stakeholder organizations. Thus, top-down forecast, bottom-up forecast, marketing forecast and collaborative forecast will be used to arrive at a consensus forecast. Scenario DescriptionThe different forecasts including the top-down,

54、 bottom-up, marketing, operations and sales are compared and contrasted by the various forecast owners and based on considerations such as revenue targets, life-cycle considerations and capacity a consensus forecast is determined. This is the final forecast that is used by the supply planning proces

55、s.Inputso Top down forecasts, bottom up forecasts, etc. at a specific node (intersection of product and geography) in the hierarchy.Outputso Consensus forecastBenefitso Communication between different organizations is achievedo Multiple data points can be displayed, allowing for analysis, comparison

56、s and metricso Emphasizes data analysis and reduced data gatheringI2 Products UsedTRADEMATRIX Demand PlannerAttach-Rate Forecasting/Dependent Demand Forecasting in Configure-to-Order environmentsDefinition In a Configure To Order (CTO) manufacturing environment, a particular product model can be sol

57、d with several options. The customer chooses the exact configuration at the time of placing an order. However, for the purpose of procuring these parts, the enterprise will need to forecast the mix of options that will potentially be sold. The forecast percentage mix of options is called “attach rat

58、es”. The consensus process essentially determines the forecast at the product model level. This process performs the option mix analysis to forecast attach rates. The attach rates can be varying by time and/or geography. Product or Product-series level forecasts will be broken down into the components or options that comprise them by using attach rates. Attach rates can be manually input or fore