(12) (10) Patent N0.: US 7,043,320 B1 Roumeliotis et a]. (45) Date of Patent: May 9, 2006

Transcription

1 United States Patent US B1 (12) (10) Patent N0.: Roumeiotis et a]. (45) Date of Patent: May 9, 2006 (54) METHOD AND APPARATUS FOR PLANNING 5,369,570 A * 11/1994 Parad /8 A MANUFACTURING SCHEDULE USING AN 5,402,350 A * 3/1995 Kine /101 ADAPTIVE LEARNING PROCESS 5,737,728 A * 4/1998 Sisey et a /8 6,233,493 B1* 5/2001 Cherneif et a /95 (75) Inventors: George Roumeiotis, Meno Park, CA * 7/2002 crampton et a1 700/100 (Us) J th L. K. ht S 6,490,566 B1 * 12/2002 Schmdt /8 ma 0 0" mg an 6,591,153 B1 * 7/2003 Crampton et a1. 700/103 Mateo CA (Us) 2004/ A1* 2/2004 Crampton et a /101 (73) Assignee: JRG Software, Inc., San Mateo, CA * cited by examiner US ( ) Primary ExamineriLeo Picard ( * ) Notice: Subject to any discaimer, the term of this patent is extended or adjusted under 35 Assistant ExamineriRyar Jarrett _ (74) Attorney, Agent, or Frmiwhite & Case LLP U.S.C. 154 b b 0 d. ( ) y ays (57) ABSTRACT (21) App. N0.: 10/846,204 _ An apparatus and method for a production panning system (22) Fedi May 14, 2004 using a process of adaptive earning is discosed. A produc tion panning system receives production objects Wherein Reated U-s- APPhcatmn Data each production object has at east one attribute. In one (63) Continuation of appication NO 10/687,032,?ed on embodiment, the production objects incude information Oct now abandoned reating to product to be manufactured. After receipt of s 3 production objects, mutipe preference scores according to (51) Int_ C]_ the attributes of the production objects are cacuated. In one G06F 19/00 (200601) embodiment, each preference score represents desirabiity of (52) us. C / /99 transition from manufacturing a?rst Object to a Second (58) Fied of Cassi?cation Search /97 Obie The System Subsequenty identi?es a Suggested Pro 700/ duction pan Which incudes a sequence order of two or more See appication?e for com pete searéh hi story objects in response to the preference scores. In one embodi ' ment, the panning preferences are updated or modi?ed by (56) References Cited an adaptive earning system in response to adjustments of the suggested production pant by a panner. US. PATENT DOCUMENTS 4,931,944 A * 6/1990 Richter et a / Caims, 10 Drawing Sheets AAQ Scenario Description $.41 Panning Preferences From Panner Decision Engine 541 Mi Suggested Pan 5.52 Panning Preferences From Historica Data Mi Panner Adjustments Fina Pan M

2 U.S. Patent May 9, 2006 Sheet 1 0f 10 Scenario Description 432 v Decision Engine Panning Preferences From Panner _4_ % Panning Preferences 1 From Historica Data Suggested Pan Q _gg L Panner Adjustments i v Fina Pan ii FIG 1

3 U.S. Patent May 9, 2006 Sheet 2 0f 10 b 0) 0 Scenario Description Decision Engine 48 Suggested Pan Panning Preferences From Panner 5% Panning Preferences From Historica Data m Adaptive Learning System?g V Panner Adjustments Fina Pan FIG 2

4 U.S. Patent May 9, 2006 Sheet 3 0f 10 Scenario Description 42 Decision Engine b 4 00 Suggested Pan Panning Preferences From Panner 415 Panning Preferences From Historica Data 446 Adaptive Learning System 162 V Panner Adjustments 52 Operationa Adaptive Learning FIG 3

5 U.S. Patent May 9, 2006 Sheet 4 0f 10 Production - Requirements Historica 401 Data i v Optmzer Learning.. Adaptive Suggested Pan 404 i Adjustment Pianner i Fina Pian Manufacturing Fm FIG 4

6 U.S. Patent May 9, 2006 Sheet 5 0f Receiving Production Objects 502 Cacuating a Puraity of Preference Scores 504 Identifying Suggested Production Pan 506

7 U.S. Patent May 9, 2006 Sheet 6 0f 10 0) 0 Identifying Attributes 602 Obtaining Transition Preferences 606 Cacuating Preference Scores 608 Providing Panning Preferences 610 Receiving A Fina Production Pan 612 Modifying Panning Preferences 614 FIG 6

8 U.S. Patent May 9, 2006 Sheet 7 0f 10 m tr i => j Historica 702 Data 703 Ai => A Average (dadua) 710 Statistica Expected Average (drandom) 712 random 1) Historica Transition Preferences 716 FIGT

9 U.S. Patent May 9, 2006 Sheet 8 0f 10 r 1 Launch Production Panner Appication 302 V Log-in to Production Panner Appication 304 Provide Task Information 306 Obtain data from Adaptive Learning Unit 308 i Create Schedue for Performing Task 310 Save and Compare Different Manufacturing Schedues to Create Desirabe Pans 312 Output the Schedue 314 r 1 FIG 8

10 U.S. Patent May 9, 2006 Sheet 9 0f 10 _QQ Processor Main Mem Bus 211 Static Bus Producj?on Panning Memory Contro Modue Z 1_ () 212 Dispay Keyboard Cursor Communication Contro Device Devce r A

11 U.S. Patent May 9, 2006 Sheet 10 0f 10 Cient Cient Suppier Storage 122 V?de Area Network 102 Loca Area Network 106 Panning System Adaptive Learning 500 Cient 114 Cient 116 FIG 10

12 1 METHOD AND APPARATUS FOR PLANNING A MANUFACTURING SCHEDULE USING AN ADAPTIVE LEARNING PROCESS CROSS REFERENCE TO RELATED APPLICATION This appication is a continuation of appication Ser. No. 10/687,032,?ed Oct. 16, 2003, now abandoned. This appication is reated to the foowing co-pending appica tion, assigned to the assignee of the present invention. Appication Ser. No. 10/845,921,?ed May 14, 2004, entited Method and Apparatus for Panning Manufacturing Schedue Using a Graphica Panning Board, which is a continuation of appication Ser. No. 10/611,561,?ed Jun. 30, 2003, now abandoned. 2 embodiment, each preference score represents desirabiity of transition from manufacturing a?rst object to a second object. The system subsequenty identi?es a suggested pro duction pan which incudes a sequence order of two or more objects in response to the preference scores. In one embodi ment, the panning preferences are updated or modi?ed by an adaptive earning system in response to adjustments of the suggested production pant by a panner. Accordingy, the system is capabe of earning the panning practices through the panning process rather than requiring a pro grammatic description of panning practices. The earned panning preferences are then used as guidance to create an optima production pan. Additiona features and bene?ts of the present invention wi become apparent from the detaied description,?gures and caims set forth beow. FIELD OF THE INVENTION BRIEF DESCRIPTION OF THE DRAWINGS The present invention reates to apparatus and methods for information processing. More speci?cay, the present invention reates to a process of panning manufacturing schedues using an adaptive earning process. BACKGROUND Increasing productivity, owering manufacturing costs, and improving on-time performance are common concerns for most manufacturing pants and factories. With increasing market pressure for rapid reaction to?uctuating demand, scheduing an e?icient manufacturing process, which maxi mizes the utiization of various resources, and minimizes equipment changeovers and downtimes, has become increasingy chaenging. Historicay, probems reating to the panning of manufacturing schedues are resoved by skifu technicians and panners. The panning process of a manufacturing schedue typicay requires a skifu panner to make various adjustments to the panning preferences to anticipate?uctuations in demand and/or unexpected events such as equipment faiure and abor issues. In the past, the production pan for a factory has typicay been generated by a panner who heuristicay appies his panning preferences regarding desirabe grouping, etc. of the required production tasks. These panning preferences have usuay been acquired by the panner during many years of experience at the factory. Such panning preferences are often vita inteectua property of a manufacturing company. A probem is that this critica know-how may be absent when the skifu panner is absent. Another probem is that, with the increasing compexity and faster pace of production panning, the skifu panner has a more di?icut time to quicky and consistenty appy and/or adjust the panning preferences. Accordingy, there is a need in the art to quantify the adjustments of panning preferences to improve the panning process for manufacturing schedues. SUMMARY OF THE INVENTION An apparatus and method for a production panner using a process of adaptive earning is discosed. A production panning system receives production objects wherein each production object has at east one attribute. In one embodi ment, the production objects incude information reating to the product to be manufactured. After receipt of production objects, mutipe preference scores according to the attributes of the production objects are cacuated. In one The present invention wi be understood more fuy from the detaied description given beow and from the accom panying drawings of various embodiments of the invention, which, however, shoud not be taken to imit the invention to the speci?c embodiments, but are for expanation and understanding ony. FIG. 1 is a?owchart iustrating a production panning system using panning preferences in accordance with one embodiment of the present invention; FIG. 2 is a?owchart iustrating a production panning system using an adaptive earning in accordance with one embodiment of the present invention; FIG. 3 is a?owchart iustrating a production panning system having an operationa adaptive earning in accor dance with one embodiment of the present invention; FIG. 4 is a bock diagram iustrating a process of adaptive earning in accordance with one embodiment of the present invention; FIG. 5 is a?owchart iustrating a process of adaptive earning in accordance with one embodiment of the present invention; FIG. 6 is a?owchart showing various steps to impement a process of adaptive earning in accordance with one embodiment of the present invention; FIG. 7 is a?owchart iustrating an exampe of cacuating a transition preference vaue in accordance with one embodiment of the present invention. FIG. 8 iustrates a computer system in which the pro duction panner may be impemented in accordance with one embodiment of the present invention; FIG. 9 is a?owchart iustrating a process of creating and modifying a manufacturing schedue using the production panner in accordance with one embodiment of the present invention; and FIG. 10 is a computer network topoogy iustrating a network environment in which the production panner can be impemented in accordance with one embodiment of the present invention. DETAILED DESCRIPTION In the foowing description, for purposes of expanation, numerous speci?c detais are set forth to provide a thorough understanding of the present invention. It wi be apparent, however, to one skied in the art that these speci?c detais may not be required to practice the present invention. In

13 3 other instances, we-known circuits and devices are shown in bock diagram form to avoid obscuring the present invention. In the foowing description of the embodiments, substan tiay the same parts are denoted by the same reference numeras. A production panning system for providing manufactur ing schedues using panning preferences is discosed. In one aspect, the production panning system generates a manu facturing schedue that incudes a preferred ordering of tasks to be manufactured by a production ine. The production panning system, in one embodiment, identi?es mutipe attributes associated with each task. Attributes identify char acteristics of an item that in?uences production panning decisions. For exampe, the attribute of Coor may have possibe vaues ight, medium, and dark. Upon determina tion of the attributes, the production panning system com putes preference scores. A preference score represents a panning preference for switching from manufacturing a?rst task to producing a second task. In one embodiment, the panning preferences are updated or modi?ed by an adaptive earning system. The adaptive earning of panning preferences incudes anayzing the decisions of an expert panner and modifying panning preferences used by a decision engine. The deci sion engine generates a suggested pan or schedue accord ing to the panning preferences. An advantage of using the adaptive earning is to capture an expert panner s know edge so that a high quaity production pan can be automati cay generated. Another advantage of using adaptive eam ing is to assist a ess experienced panner in producing a high quaity pan. A skifu panner can aso bene?t from using the adaptive earning during time critica situations because the adaptive earning enhances the quaity of the suggested pan. In another embodiment, the adaptive earning system monitors mutipe forecasts from various panners. The forecasts invove in actua demand given historica patterns and current forecasts from different parts of the organization such as saes, marketing, and manufacturing. The system then uses the forecasts to provide a best forecast of the actua demand that mimics the decision processes of the best panner. FIG. 1 is a?owchart iustrating a production panning system 440 using panning preferences in accordance with one embodiment of the present invention. At bock 442, the process receives a scenario description. In one embodiment, the scenario description incudes production requirements. The scenario description incudes information reevant to the task such as demand for goods, production rates, capacities, changeover durations, etc. At bock 444, a panner provides panning preferences to a decision engine. The panner may input panning prefer ences from his or her own experience. In another aspect, the panner may use a digita processing device to identify preferred panning preferences. At bock 446, the process obtains panning preferences from a historica data storage. The data storage may be ocated in a host computer or a remote system. At bock 448, a decision engine receives data from bock 442, 444, or 446. In another embodiment, the decision engine receives the data from one of the three bocks 442*446 at any given time. At bock 450, the process generates a suggested pan in accordance with the panning preferences. At bock 452, the process aows a panner to manuay adjust the suggested pan as he or she determines is appro priate At bock 454, the process generates a?na pan. The?na pan, in one embodiment, is forwarded to a production?oor for execution. In one embodiment, the?na pan is the preferred production schedue for manufacturing objects in a sequence under the avaiabe information to the production panning system. FIG. 2 is a?owchart iustrating a production panning system 460 using an adaptive earning in accordance with one embodiment of the present invention. The?owchart iustrated in FIG. 2 is simiar to the?owchart iustrated in FIG. 1 except bock 462. At bock 462, an adaptive earning system receives the?na pan. In one aspect of the present invention, an adaptive earning system is capabe of captur ing and anayzing decisions made by an expert panner. In genera, terms of the adaptive earning system, adaptive earning of panning preferences, and adaptive eaming may be used interchangeaby herein. In one embodiment, the adaptive earning system is used to generate an automatic production pan in accordance with the production require ments or considerations. The production requirements or considerations incude ingredients, attributes, grouping, etc. In another embodiment, the adaptive earning system is used to assist a ess experienced panner to create a cost-effective and e?icient production pan. In yet another embodiment, the adaptive earning system enabes consistent appication of panning preferences for decision support to generate a high-quaity production pan. The adaptive earning system is aso capabe of anayzing the historica data by pacing ess emphasis on data from the distant past as compared with historica data from the recent past. In one embodiment, the adaptive earning system is appied in the area of suppy chain management. The suppy chain management incudes production scheduing, materia procurement, sourcing, inventory pacement, avaiabe-to promise, truck oading and routing, promotions, and et cetera. The terms of production scheduing and production panning may be used interchangeaby herein. The suppy chain management of materia procurement, in one embodiment, is to generate a procurement pan to meet production requirements. The scenario description incudes data eements and/or parameters for generating a procurement pan. For exampe, a seection of a vendor may be in?uenced by the parameters of materia, quantity, price, quaity and reiabiity. The vendor may change from month to month depending on various business factors such as vendor terms and materia inventory. The suppy chain management of sourcing, in another embodiment, is to generate a sourcing pan to fu? cus tomer demand either from existing inventory or from new production. The scenario description may incude custom ers, demands, inventory positions, factory capacities, trans portation costs, etc. The suppy chain management of inventory pacement generates an inventory pan that speci?es the inventories to pace at distribution centers, factory warehouses, etc. The scenario description may incude?rm and forecasted demand, voatiity of demand, characteristics of inventory ocations, etc. The suppy chain management of avaiabe-to-promise is con?gured to respond to various saes inquiries. The sce nario description may incude cass of customer,?rm and forecasted demand, voatiity of demand, inventory posi tions, factory capacities, etc. The character of the avaiabe to-promise is accuracy and rea-time response. The suppy chain management of truck oading and routing, in one aspect, provides a pan for oading a truck and routing the truck for customer deiveries. The scenario

14 5 description may incude the customer orders, characteristics of the trucks such as weight and size constraints, routes and distances, deivery schedues, etc. The suppy chain management of promotions generates a pan for production eves and pricing of promotiona items. The scenario description may incude historica campaign effectiveness, factory capacities, etc. In one embodiment, a production panning system, which is used as a decision support system for suppy chain management, incudes a decision engine, wherein input of the decision engine is data eements of the scenario descrip tion and output of the decision engine is a suggested pan. The decision engine, in one embodiment, provides a map ping between the scenario description and the suggested pan. The mapping invoves panning preferences, which are parameters that determine the behavior of the decision engine. After the decision engine generates a suggested pan, expert panners may modify the pan according to their past experience. For exampe, in a procurement decision support system, athough the decision engine may suggest vendor A given the scenario description, the expert panner may change vendor A to vendor B for various reasons such as recent?nancia hardships of vendor A. Aso, an expert panner, for exampe, may have heard that vendora is facing abor disputes and therefore presents a greater business risk than vendor B. The adjusted pan becomes the?na pro curement pan. The adaptive earning system observes and anayzes the scenario description and the corresponding?na pan. The adaptive earning system modi?es the parameters within the decision engine to produce a more efficient production schedue. The panning preferences are modi?ed so that a more e?icient production pan may be generated for simiar scenario descriptions in the future. FIG. 3 is a?owchart iustrating a production panning system 470 having an operationa adaptive earning in accordance with one embodiment of the present invention. The?owchart iustrated in FIG. 2 is simiar to the?owchart iustrated in FIG. 3 except bock 472. At bock 472, an operationa adaptive earning aows an evauator to objec tivey evauate actua resuts from the production ine against the panning preferences. The evauator may update the panning preferences based on the actua resuts of the production ine. FIG. 4 is a bock diagram of production panning system 400 having an adaptive earning system in accordance with one embodiment of the present invention. In one embodi ment, production panning system 400 is expressed in com puter executabe instructions, which causes a genera pur pose or specia purpose computer to perform the steps set forth in the production panning system 400. Aternativey, the steps of production panning system 400 may be per formed by speci?c hardware components, or by a combina tion of computer programming code and custom hardware components. At bock 401, the process obtains production require ments, which incudes objects or goods, time imitations, ingredients, etc. In one embodiment, the process identi?es the origin of production requirements, which may come from a oca system or a remote system through the Internet. Upon receipt of production requirements, the process pro ceeds to bock 402. At bock 402, the process obtains panning preferences from an adaptive earning system in accordance with the production requirements. In one embodiment, the adaptive earning system identi?es and fetches reevant historica panning data according to the production requirements. Historica data may be stored in a oca database or in a remote database. At bock 404, the process receives various parameters incuding production requirements, panning preferences, etc. A suggested pan is generated in accordance with the various parameters. At bock 406, the process aows a panner 430 to adjust the suggested pan. In one embodiment, an experienced panner 430 can manuay change certain parameters to improve the production pan. For exampe, the panner 430 may know the recent abor issues with a particuar suppier and it is desirabe to avoid these risks in an effective production pan. At bock 408, upon competion of adjustment by a pan ner, the process provides a?na production pan. The?na production pan or schedue is ready to be executed at a manufacturing production ine. In one embodiment, the manufacturing?oor may continuousy receive rea-time updates from the production panning system to improve the e?iciency of the production process. In one embodiment, the?na pan is forwarded to the adaptive earning system for updating panning preferences in accordance with the?na production pan. At bock 420, the process evauates the?na production pan and captures adjustments made by the panner. Some historica data may be updated according to the?na pro duction pan. The adaptive earning system 420 aso pro vides parameters such as panning preferences to the deci sion support optimizer 402. It shoud be obvious to one skied in the art that it does not depart from the scope of the present invention if additiona bocks are added to or sub tracted from the process 600. FIG. 5 is a?owchart 500 iustrating a process of adaptive earning in accordance with one embodiment of the present invention. At bock 502, the process receives production objects. Production objects, in one embodiment, incude at east a portion of the production requirements such as tasks, time constraints, and other characteristics of each task. The information about production objects may be obtained from the host computer and/or from a remote computer via the Internet. At bock 504, the process cacuates a puraity of pref erence scores according to various parameters such as transitiona preference vaues. In one embodiment, a statis tica agorithm is used to cacuate the preference scores. In another embodiment, the process is capabe of providing preference scores for two adjacent production tasks. In yet another embodiment, the process is abe to provide prefer ence scores for a sequence of production tasks. At bock 506, the process identi?es and provides a sug gested production pan in response to the preference scores and panning preferences. It shoud be obvious to one skied in the art that it does not depart from the scope of the present invention if additiona bocks are added to or subtracted from the process 500. The adaptive earning system is a part of the production panning system that generates e? ective production sched ues or production pans. In one embodiment, a production pan speci?es a sequence in which the factory shoud execute in order to meet the production requirements. For exampe, a factory foows various manufacturing opera tions in accordance with production requirements (or pro duction tasks). Tabe 1 iustrates severa parameters:

15 SKU P1 P3 P6 P7 TABLE 1 Quantity 1400 Cases 4700 Cases 2500 Cases 3600 Cases Wherein SKU refers to Stock Keeping Unit, Which is generay a unique identi?er for each item. If the factory has one production ine, an exempary production pan is ius trated in Tabe 2. TABLE 2 Line Start SKU Quantity L1 06:00 P Cases L1 09:30 P Cases L1 13:00 P Cases L1 15:00 P Cases To create an e?icient production pan, various consider ations need to be baanced. One consideration is to group the production items that share a substantiay simiar charac teristic. For exampe, it may be desirabe to group the production of items that share simiar ingredients since switching ingredients requires ceaning and recon?guring the production ine. Aso, it takes time and abor to switch from one ingredient(s) to another ingredient(s). Another consideration may be to group the production of items that require the same speciaized abor, since under-utiizing such abor can incur unnecessary costs. Other considerations may invove seecting the production ines, time imitations, etc. FIG. 6 is a?owchart having more detaied steps of an adaptive earning process 600 in accordance With one embodiment of the present invention. At bock 602, the process identi?es a puraity of attributes associated With production objects or tasks. In one embodiment, the pro duction objects are identi?ed in the production require ments. As mentioned earier, the panning preferences reated to grouping of production tasks can be quanti?ed using the concept of attribute. Attributes are characteristics of an object or item that in?uences production panning decisions. An experienced panner concerns a particuar set of attributes that may be unique to his or her manufacturing operations. For exampe, the attributes for roasting cofee may be Roast, Package and Channe. Each e?fective pro duction pan may contain a sequence of production tasks so that tasks With simiar attributes are grouped together to improve the throughput and cost-effectiveness of the manu facturing operation. At bock 606, the process identi?es and obtains transition preference vaues in accordance With the puraity of the attributes. The panning preferences associated With a given attribute can be quanti?ed by specifying transition prefer ences. The term transition preferences and transition pref erence vaues may be used interchangeaby herein. For exampe, attributeahas the possibe vaues A1, A2,...,AN. The transition preference for switching from producing an item With attribute vaue A. to producing an item With attribute vaue Aj is denoted by TP(A-:>Aj). For exampe, a higher vaue of TP(A-:>Aj) indicates a stronger panning preference for scheduing production of an item With the attribute vaue Aj immediatey after production of an item With the attribute vaue Ai. In one embodiment, the transition preferences for an attribute can be speci?ed by the panner, or they can be computed from historica panning data. At bock 608, the process cacuates a puraity of pref erence scores for two adjacent production tasks and/or a sequence of production tasks. The preference score for switching from producing a?rst item to producing a second item is computed by adding the transition preference for each attribute. For exampe,?rst item I 1 has attribute vaues: Attribute AIA Attribute BIB Attribute CIC And, second item I2 has attribute vaues: Attribute AIA2 Attribute BIB2 Attribute CIC2 Then the preference score PS(I1:>I2) for switching from producing the?rst item I2 to producing the second item I2 is: The preference score for a sequence of production tasks is computed by adding together the preference scores for adjacent production tasks in a sequence. For exampe, a sequence of production tasks invoves production of a?rst item I1, foowed by production of a second item I2, foowed by production of a third item I3, etc. The equation of preference score PS(I1:>I2:>I3:>... ) is: PS(I1:>I2 :>I3:>... ):PS(I1:>I2)+PS(I2:>I3)+PS (I3:>I4)+... At bock 610, the process evauates the preference scores and production considerations. In one embodiment, a higher vaue of a preference score indicates a stronger panning preference for scheduing production. The process provides panning preferences according to the puraity of preference scores. At bock 612, the process receives a?na production pan. In one embodiment, the adaptive earning system faciitates continuous feedback of the panning process to enhance the effectiveness of the production pan. In another embodiment, the adaptive earning system enabes the production pan ning system to continuousy earn. At bock 614, the process modi?es the panning prefer ences according to the?na production pan. In one embodi ment, the process captures the knowedge of an experienced panner by identifying the adjustments made by the panner and updating the panning preferences in accordance With the adjustments. In one embodiment, the adaptive earning system updates the panning preferences When it is invoked by an operator or system. In another embodiment, the adaptive earning system updates the panning preferences at pre-de?ned times. In yet another embodiment, the adaptive earning system updates the panning preferences on speci?ed events eg When a pan is forwarded for execution. It shoud be obvious to one skied in the art that it does not depart from the scope of the present invention if additiona bocks are added to or subtracted from the process 600. FIG. 7 is a?owchart iustrating an exampe of data processing 700 for cacuating a transition preference vaue in accordance With one embodiment of the present inven tion. At bock 702, the process identi?es?rst item I1- and second item I], Wherein item is the object to be manufac tured. In one embodiment, I1- and Ij are derived from pro duction requirements. At bock 703, the process identi?es and fetches reevant parameters from historica data storage in response to II- and

16 Ij. The reevant parameters incude information such as attributes and panning preferences. In one embodiment, the historica data storage is ocated in a oca system. In another embodiment, the historica data storage is ocated in a remote system. At bock 704, the process receives attributes A- and A] associated With II. and I], respectivey. A. and A]. are subse quenty forwarded to bock 710 and 712. At bock 710, the process obtains an average transition preference vaue for A. and Aj. For exampe, the distance between two attribute vaues A- and Aj in the sequence may be the difference in the indexes of those attribute vaues. dam ; is referred to the average distance from an occurrence of the attribute vaue A- in the sequence to the subsequent occurrence of the attribute vaue Aj. For exampe, if every occurrence of A- is immediatey foowed by an occurrence of A], then dacmaf. At bock 712, the process obtains a statisticay expected transition preference vaue for A- and Aj. Let dmndom be the statisticay expected average distance if the attribute vaues in the sequence are re-distributed to form a randomized sequence. In one embodiment, dmndom can be computed by performing randomizations and then the process identi?es the mean of dam ; obtained over a randomizations. At bock 714, the process computes the transition pref erence vaue using the formua: TP(A-:>Aj):exp[-(dacmaF )/ (dmn dam-1)], Wherein the exp[ ] represents the exponen tia function. At bock 716, the process receives the historica transition preference vaues from bock 714. These historica transition preferences can then be provided as parameters to decision support systems, such as a system to automaticay generate the preferred sequence. In one aspect, the adaptive earning of transition preferences can be accompished by automati cay updating the historica transition preferences upon the occurrence of a triggering event. The production panner having an adaptive earning mod ue, in one embodiment, is impemented in a computer program, Which can be oaded to a genera-purpose com puter to perform the process of panning of manufacturing schedues. The production panner aows a panner to expicity express preferences and outputs a manufacturing schedue based on the panner s preferences. The prefer ences may incude equipment preferences, shift preferences, sequencing preferences, and desired grouping of tasks With simiar attributes. In another embodiment, the production panner is impemented in an integrated circuit (IC). In yet another embodiment, the production panner is impemented in a combination of IC and programming code. FIG. 8 is a?owchart iustrating a process of creating and modifying manufacturing schedues using production pan ner 300 in accordance With one embodiment of the present invention. At bock 302, the process aunches an appication of production panner 300. In one embodiment, the appi cation is activated through a remote cient system connected through the Internet. In another embodiment, the appication of production panner 300 is aunched by a cient system Within a oca-area network. In yet another embodiment, the appication of production panner 300 is aunched by the system that contains production panner 300. At bock 304, the process activates a ogin mechanism. The ogin mechanism requires a user to input or sign-in his or her account information. Account information, such as user identi?er, password, et cetera, contains information associated With the user or cient. Upon verifying the account information, the process moves to the next bock At bock 306, the user or panner suppies task informa tion, Which is aso known as production requirements. In one embodiment, the task information is obtained directy from another system Without user intervention. The task informa tion identi?es types of products or objects to be manufac tured. The task information further incudes quantities of objects or products to be manufactured, required shipment dates, production ines and equipment assigned to manufac ture the products, changeover requirements, et cetera. Changeover refers to the activities necessary to convert a production ine from manufacturing a?rst product to a second product. At bock 308, the process obtains attributes according to the task information. After identi?cation of the attributes, the process obtains preference scores according to the attributes. The process uses preference scores to identify a puraity of panning preferences. In one embodiment, adaptive earning modue retrieves some data from a past panning database. Upon the process obtains the data from adaptive earning modue, the process moves to the next step. At bock 310, the process creates a suggested manufac turing schedue in response to the data obtained from the adaptive earning modue. In one embodiment, production panner retrieves various contro information such as the user s desirabe preferences, dispays a schedue for per forming the task and aso aows the user to modify the schedue. In one embodiment, generation and modi?cation of the schedue can be interactivey performed on a graphi ca panning board. At bock 312, the user can save and compare different manufacturing schedues to create one or more desirabe pans. Production panner 300 aows a user or a skifu panning expert to ater the manufacturing schedues guided by instantaneous messages or Warnings resuting from the user s movement of task(s). At bock 314, the process outputs a manufacturing sched ue, Which the user concudes is the most desirabe pan under the condition of avaiabe data. The process proceeds to bock 312 if the user Wants to further re?ne the schedue. To impement these steps, production panner 300 may incude various commerciay avaiabe components. In one embodiment, production panner 300 incudes Java Appets, JSPs, Servets, Standard Tag Libraries, Java components, and SQL database tabe for impementing and creating manufacturing schedues. In this embodiment, Java Appets are used to improve user interactions With the cient browser. Java components incude simpe beans, hepers, and/or business services. Servets are used to make initia contact for handing requests from the cient. Standard Tag Libraries are needed to minimize scriptet code inside of the JSPs. SQL database tabe is used for data storage. It shoud be obvious to one skied in the art that it does not depart from the scope of the present invention if additiona components are added to or subtracted from the above-described com ponents. FIG. 9 iustrates an exampe of a computer system 200 in Which the features of the present invention may be impe mented. Process of the production panner can be impe mented in any processor-based computer system, such as a PC, a Workstation, or a mainframe computer. It Wi be apparent to those of ordinary ski in the art that other aternative computer system architectures may aso be empoyed. It shoud aso be noted that the terms manufac turing scheduing, manufacturing panning, manufacturing scheduing and panning, production panning, production scheduing, and production panning and scheduing coud be used interchangeaby hereinafter.

17 11 Referring back to FIG. 9, computer system 200 incudes a processing unit 201, an interface bus 211, and an input/ output ( IO ) unit 220. Processing unit 201 incudes a processor 202, a main memory 204, a system bus 211, a static memory device 206, a bus contro unit 205, a mass storage memory 207, and a production panning modue 230. Bus 211 is used to transmit information between various components and processor 202 for data processing. Processor 202 may be any of a Wide variety of genera purpose processors or microprocessors such as PentiumTM microprocessor, MotoroaTM 68040, or PoWer PCTM micro processor. Main memory 204, Which may incude mutipe eves of cache memories, stores frequenty used data and instruc tions. Main memory 204 may be RAM (random access memory), MRAM (magnetic RAM), or?ash memory. Static memory 206 may be a ROM (read-ony memory), Which is couped to bus 211, for storing static information and/or instructions. Bus contro unit 205 is couped to buses 211*212 and contros Which component, such as main memory 204 or processor 202, can use the bus. Bus contro unit 205 manages the communications between bus 211 and bus 212. Mass storage memory 207, Which may be a magnetic disk, an optica disk, hard disk drive,?oppy disk, CD-ROM, and/or?ash memories for storing arge amounts of data. Production panning modue 230, Which incudes an adaptive earning modue, may in one embodiment, be an independent component (IC) that performs functions of panning manufacturing schedues. In another embodiment, production panning modue 230 may reside Within the processor 202, main memory 204, and/or static memory 206. I/O unit 220, in one embodiment, incudes a dispay 221, keyboard 222, cursor contro device 223, and communica tion device 225. Dispay device 221 may be a iquid crysta device, cathode ray tube ( CRT ), touch-screen dispay, or other suitabe dispay device. Dispay 221 projects or dis pays images of a graphica panning board. Keyboard 222 may be a conventiona aphanumeric input device for com municating information between computer system 200 and computer operator(s). Another type of user input device is cursor contro device 223, such as a conventiona mouse, touch mouse, trackba, or other type of cursor for commu nicating information between system 200 and user(s). Communication device 225 is couped to bus 211 for accessing information from remote computers or servers, such as server 104 or other computers, through Wide-area network 102. Communication device 225 may incude a modem or a network interface device, or other simiar devices that faciitate communication between computer 200 and the network. Computer system 200 may be couped to a number of servers 104 via a network infrastructure. The present invention incudes various processing steps, Which Wi be described beow. The steps of the present invention may be embodied in machine or computer execut abe instructions. The instructions can be used to cause a genera purpose or specia purpose system, Which is pro grammed With the instructions to perform the steps of the present invention. Aternativey, the steps of the present invention may be performed by speci?c hardware compo nents that contain hard-wired ogic for performing the steps, or by any combination of programmed computer compo nents and custom hardware components. Whie embodi ments of the present invention Wi be described With reference to the Internet, the method and apparatus described herein is equay appicabe to other network infrastructures or other data communications environments FIG. 10 is a computer network topoogy 100 iustrating a network environment in Which the production panner can be impemented in accordance With one embodiment of the present invention. In this network environment, a panning system 130 is couped to a Wide-area network 102. Wide area network 102 incudes the Internet, or other proprietary networks incuding America On-LineTM, SBCTM, Microsoft NetWorkTM, and ProdigyTM. Wide-area network 102 may further incude network backbones, ong-hau teephone ines, Internet service providers, various eves of network routers, and other means for routing data between comput ers. Panning system incudes the production panner, Which may be impemented either in hardware or software or a combination of hardware and software. Panning system 130 incudes an adaptive earning modue 500. Panning system 130 is capabe of accepting user s input during the process of panning a manufacturing schedue. Server 104 is couped to Wide-area network 102 and it is, in one aspect, used to route data to cients 114*116 through a oca-area network ( LAN ) 106. The LAN connection aows cient systems 114*116 to communicate With each other through LAN 106 and to communicate With panning system 130 via LAN 106 and server 104. Using conven tiona network protocos, panning system 130 may com municate through Wide-area network 102 to a puraity of cient computer systems 110*112, suppier system 120 and storage device 122. For exampe, cient system 110 is connected directy to Wide-area network 102 through direct or dia-up teephone or other network transmission ines. Aternativey, cients 110*112 may be connected through Wide-area network 102 using a modem poo. Using one of a variety of network connection means, panning system 130, Which coud be a persona computer ( PC ), a mini-computer, a server, a Workstation, or a mainframe computer, can support mutipe appications of production panner to cients across the network. In one embodiment, panning system 130 may store and retrieve various eectronic information (or data) in storage system 122 through Wide-area network 102. Panning system 130 is capabe of obtaining various information such as suppier s inventory from suppier system 120 via the network. It shoud be obvious to one skied in the art that it is Within the scope of the present invention if additiona systems are added to or subtracted from the computer network 100. In the foregoing speci?cation the invention has been described With reference to speci?c exempary embodiments thereof. It Wi, however, be evident that various modi?ca tions and changes may be made thereto Without departing from the broader scope of the invention. The speci?cation and drawings are, accordingy, to be regarded in an ius trative rather than restrictive sense. What is caimed is: 1. A computer-impemented method of production pan ning comprising: receiving production objects, Wherein each object has at east one attribute; identifying one or more transition preference vaues asso ciated With each of the attributes; cacuating a puraity of preference scores according to the attributes of the production objects and said one or more transition preference vaues, Wherein each pref erence score represents desirabiity of transition from manufacturing a?rst object to a second object; generating a suggested production pan Which incudes a sequence order of two or more production objects in response to the preference scores; receiving user adjustments to the suggested production pan;

18 13 adjusting at east one of the transition preference vaues in response to said user adjustments to the suggested production pan; and storing at east one of the transition preference vaues in a database. 2. The computer-impemented method of caim 1, Wherein cacuating a puraity of preference scores further incudes: retrieving one or more statistica agorithms from a com puter-readabe medium; and utiizing said one or more statistica agorithms to com pute transition preference vaues. 3. The computer-impemented method of caim 1, Wherein cacuating a puraity of preference scores further incudes identifying a portion of the preference scores that represents desirabiity of transition from manufacturing the?rst object to a third object. 4. A computer-impemented method of adaptive earning comprising: identifying a puraity of attributes in response to produc tion tasks; obtaining a puraity of transition preference vaues from a computer-readabe medium in response to the pura ity of attributes; cacuating a puraity of preference scores in response to the puraity of transition preference vaues; providing the puraity of preference scores to a decision engine; generating?rst production pan; receiving user adjustments to the?rst production pan; adjusting at east one of the puraity of transition pref erence vaues in response to said user adjustments; and storing said transition preference vaues in said computer readabe medium. 5. The computer-impemented method of caim 4, Wherein the step of identifying a puraity of attributes further incudes identifying a puraity of attributes associ ated With a production object. 6. The computer-impemented method of caim 4, Wherein the step of identifying a puraity of attributes further incudes obtaining a puraity of production consid erations from a production objects. 7. The computer-impemented method of caim 6, Wherein the puraity of production consideration further incudes identifying demand of goods to be produced. 8. The computer-impemented method of caim 6, Wherein the puraity of production considerations further incudes identifying changeover durations between the pro duction objects. 9. The computer-impemented method of caim 6, Wherein the puraity of production considerations further incudes identifying production rates and capacities to pro duce the required production objects. 10. The computer-impemented method of caim 4, Wherein the step of obtaining a puraity of transition pref erence vaues further incudes identifying a combination of transitions between production objects in response to the puraity of attributes. 11. The computer-impemented method of caim 4, Wherein the step of obtaining a puraity of transition pref erence vaues further incudes identifying numerica vaue representing a transition from a?rst object to a second object in response to simiarity of attribute vaues for the?rst and second objects. 12. The computer-impemented method of caim 4, Wherein the step of obtaining a puraity of transition pref erence vaues in response to the puraity of attributes further incudes cassifying a panning preference as recent or distant based upon the age of the past panning preference The computer-impemented method of caim 4, fur ther comprising utiizing a statistica agorithm to derive the preference score. 14. A computer-impemented method of manufacturing panning comprising: providing a scenario description to a decision engine; obtaining panning preferences from a data storage; providing the panning preferences to the decision engine; generating a suggested pan in response to the scenario description and panning preferences; receiving user adjustments to the suggested pan; adjusting at east one of the puraity of panning prefer ences in response to said user adjustments; and storing said modi?ed panning preferences in said data storage. 15. The computer-impemented method of caim 14, Wherein the step of receiving a description of production requirements incudes identifying at east one item to be manufactured. 16. The computer-impemented method of caim 14, Wherein the step of obtaining panning preferences from the adaptive earning engine further incudes identifying a pu raity of attributes associated With the production require ments. 17. The computer-impemented method of caim 16, Wherein the step of identifying a puraity of attributes associated With the production requirements further incudes obtaining a puraity of transition preference vaues associ ated With the puraity of attributes. 18. The computer-impemented method of caim 17, Wherein the step of obtaining a puraity of transition pref erence vaues associated With the puraity of attributes further incudes cacuating a puraity of panning prefer ences in response to the puraity of transition preference vaues. 19. The computer-impemented method of caim 14, Wherein the step of modifying the panning preferences in response to the?na pan further incudes storing updated panning preferences to a storage ocated across a network. 20. The computer-impemented method of caim 14, Wherein the step of obtaining panning preferences from the historica data storage further incudes obtaining panning preferences from a puraity of storages across a network of factories. 21. The computer-impemented method of caim 14, Wherein the step of modifying the panning preferences further incudes monitoring a puraity of forecasts from mutipe sources. 22. The computer-impemented method of caim 21, Wherein the step of monitoring a puraity of forecasts from mutipe sources further incudes seecting an optima fore cast in response to the panning preferences. 23. The computer-impemented method of caim 1, Wherein said step of identifying transition preference vaues further incudes retrieving transition preference vaues from a database. 24. The computer-impemented method of caim 1 further comprising: generating a second suggested production pan using said adjusted transition preference vaues, Wherein said sec ond suggested production pan is different from said?rst suggested production pan. 25. The computer-impemented method of caim 1 further comprising: generating a second suggested production pan using said adjusted transition preference vaues, Wherein said sec ond suggested production pan is substantiay simiar to said?rst suggested production pan. * * * * *