Logistics System for Automating Transportation of Goods

The present invention relates generally to computerized systems for expediting the shipping of goods in commerce. More particularly, the invention relates to a computerized logistics system for managing and integrating various aspects of order processing, order fulfillment and goods transportation and tracking.

In the past, computerized systems for expediting the shipping of goods have fallen into two rather diverse categories. At the low cost end of the spectrum have been the standalone postage meters and mail manifest systems used by small businesses to automate the package weighing and carrier manifest printing functions. At the other end of the spectrum are the mainframe computer-based systems employed by large nationwide mail order merchandisers. At both ends of the spectrum the systems have had a number of limitations.

The standalone mail manifesting systems are limited in that they are designed to automate only the shipping functions such as printing mailing label and mailing manifest by the shipping clerk or shipping department. As such, the conventional standalone system was not integrated with the customer order department or with the order fulfillment and order packaging departments. Hence, conventional standalone systems have lacked the ability to take order size, package size or time in transit into account when selecting the least cost carrier.

Large mainframe order processing systems are also limited. Due to the complexity of mainframe computer architecture and associated software systems, it is not practical to use these solutions in the small or moderate sized business environment. Mainframe-based systems often require years to develop and to customize for a particular organization's needs. Thereafter, large data processing departments are needed to maintain the system and keep it operational.

The present invention provides a high-performance, cost-effective logistics system which is readily adaptable to a wide variety of different organizations. The system is suitable for deployment on a single, standalone computer or on a computerized network comprising many computers. Among the advantages of the present system are (1) substantial reduction in freight costs; (2) a major increase in fulfillment accuracy; (3) convenient order tracking to facilitate warranty, lot and serial number tracking; (4) improved customer service; (5) a readily customizable system which can be adapted to virtually any shipping operation; (6) a robust system having a long useful life; (7) graphical user interface screens for easy training and use; and (8) greatly reduced implementation costs in a system with increased effectiveness.

As more fully described herein, the logistics management system of the invention facilitates the process of shipping goods by a shipper having a predefined set of shipping requirements via a carrier having a predefined rate structure. The system employs a multitasking operating system environment for running a plurality of computer processes substantially simultaneously. The environment has a means for interprocess communication whereby messages may be passed between the computer processes. A supervisory server, running in the operating system environment, provides registration services to connect one or more computer processes to the interprocess communication mechanism. The system further employs at least one rate server, also running in the operating system environment, substantially simultaneously with the supervisory server. The rate server or servers provide access to carrier rate structure data and also provide predefined data processing services using the carrier rate structure data in response to a predefined set of request messages. The predefined data processing services include the providing of response messages based at least in part on the carrier rate structure data. More specifically, the rate server or servers have registration means for communicating with the supervisory server to invoke the registration services of the supervisory server and thereby establish a connection to the interprocess communication means. In the presently preferred embodiment there is one rate server for each carrier (e.g., U.S. Postal Service, Federal Express, United Parcel Service, etc.) and these servers are provided with a complete knowledge base of all rate structure data and shipping rules and regulations pertaining to that carrier.

In addition to the supervisory server and one or more rate servers, the logistics managements system also includes at least one client process running in the operating system environment substantially simultaneously with the supervisory server and also with the rate server or servers. The client process has a user interface for collecting input information from a user about a desired operation and for providing output information. More specifically, the client process also has registration means for communicating with the supervisory server, to invoke the registration services of the supervisory server, and thereby establish a connection to the interprocess communication means. The client process has a preprogrammed set of rules which are reflective of a given shipper's predefined set of shipping requirements. The client process also has a processing means for using the preprogrammed set of rules and using at least a portion of the input information to issue request messages to one or more rate servers and to interpret response messages received from the rate servers in order to provide the output information. In the presently preferred embodiment the client process is preprogrammed with a knowledge base to reflect the shipping organization's rules, regulations and practices. In this way, the client process presents a familiar view of day-to-day operations, as seen by the organization's personnel who are responsible for taking orders, packaging goods and shipping goods to customers. Because the sometimes complex rules and regulations of the carriers are fully handled by the rate servers, users interacting with the client process do not need to have a full and complete understanding of the carrier's rules and regulations in order to properly ship goods in a cost-effective and timely manner.

As more fully set forth herein, the supervisory server, the rate server or servers and the client process or processes are interoperable through the interprocess communication means (a) to receive input information from a user via the user interface of the client process, (b) to use the input information to issue a request message to the rate server via the interprocess communication means, (c) to process the issued request message and thereby cause a response message to be generated by the rate server, (d) to send the response message to the client process via the interprocess communication means, and (e) to provide output information based on the response message. The output information can range from simply displaying information on a screen to the user, to printing a mailing label or manifest or to updating records in a company database.

For a more complete understanding of the invention, its objects and advantages, reference may be had to the following specification and to the accompanying drawings.

FIG. 1 illustrates an example application in which the logistics management system of the invention may be implemented;

FIG. 2 is an icon view of the plurality of program objects which comprise a presently preferred embodiment of the logistics management system;

FIG. 3A is a block diagram depicting a distributed architecture embodiment of the invention;

FIG. 3B is a block diagram illustrating a single, standalone CPU embodiment of the invention;

FIGS. 4A-4L, inclusive, represent exemplary user interface screens of the presently preferred embodiment;

FIG. 5 is a block diagram illustrating the presently preferred mechanism for implementing the client/server architecture, illustrating how multiple threads operate in the presently preferred embodiment;

FIG. 6 is a block diagram illustrating the presently preferred tree-structured client/server communications mechanism.

The logistics system of the invention serves as a management tool for the automated order processing, packaging, shipping and transportation of goods. The system is highly flexible and adaptable and thus the invention can be implemented in many forms. Therefore, in order to illustrate the principles of the invention an exemplary order processing, packaging and shipping operation will be illustrated and described. It will be understood that the invention provides a collection of building blocks or program objects which can be assembled in a variety of different ways to easily construct a logistics management system for practically any application.

Referring to FIG. 1, an exemplary application is illustrated. In FIG. 1, a networked architecture is illustrated in which a plurality of computers are interconnected by a local area network bus 20. Of course, the number of computers and the network architecture utilized are matters of design choice. The invention is not restricted in this regard and will operate on systems as small as a single standalone computer and as large as a global-wide area network.

FIG. 1 illustrates a simple system which includes an order processing station 22, a packaging station 24 and a shipping station 26. The order processing station might ordinarily include one or more computer terminals through which order entry personnel input a customer's order. The order entry terminal may be integrated with a point-of-sale terminal or cash register or it may be associated with or connected with a telephone system through which customer orders are placed. In this regard, while most order entry terminals are designed to be operated by an order processing clerk, direct order entry by the customer via computerized telecommunications equipment is also envisioned.

The order packaging station may also comprise one or more computer terminals to which a bar code scanning device 28 may be optionally attached. The scanning device would be used, for example, to scan the universal product code (UPC) of each item as it is picked from the warehouse shelves and placed into the shipping container 30.

The shipping station 26 similarly may include one or more computer terminals to which a scanning device 32, electronic scale 34 and mailing label printers 36 may be attached. Preferably, the printers are capable of printing the necessary shipping documents, bills of lading, manifests and so forth, as well as the appropriate package labeling. If desired in the alternative, the package label may be preprinted (e.g. at the packing station) and the scanning device 32 may be used to read the label and thereby automatically enter the package identifying number into the system.

The logistics management system of the invention may be implemented in software and run from a variety of different computer platforms. Preferably, at least portions of the logistics management software are installed and run on each of the computer terminals illustrated in FIG. 1. In addition, the logistics management system software may also be installed and run on other computers attached to the network, such as computer 38. As will be more fully described below, the logistics management system is also capable of interfacing with non-native computer systems, e.g., previously existing company database systems, via an external processing management system. To illustrate this, an external database 40 is depicted in FIG. 1. The database may be resident, for example, on a mini-computer or mainframe computer used to store company financial records. If necessary, the external database system may be connected via a gateway 42 to the local area network bus 20.

Client/Server Architecture

The presently preferred logistics management system is implemented using a client/server architecture and a multitasking operating system. Although the presently preferred system runs under the OS/2 operating system, use of OS/2 is not a requirement. Any multitasking operating system can be used. A multitasking operating system was selected for the preferred embodiment because it permits multiple processes (and multiple threads) to run effectively simultaneously. The multitasking operating system thus allows multiple programs to run effectively simultaneously and to communicate with each other through an interprocess communications (IPC) mechanism.

One benefit of the multitasking operating system is that the present invention allows the logistics management task to be split into multiple pieces. This architecture is quite advantageous since it allows updates or changes to be effected with respect to part of the system without affecting the rest of the system. The client/server architecture derives benefit from the multitasking operating system by allowing the overall logistics management task to be subdivided along functional lines. As will be more fully explained below, the presently preferred embodiment places carrier-related information, such as shipping rates, shipping rules, time in transit information and the like in one or more rate servers. These servers are responsible for making all determinations regarding how a given carrier's rules and rate structures are to be interpreted. The presently preferred embodiment facilitates the particular shipper's requirements, such as order taking, order fulfillment, inventory control and the like, in one or more client applications. These client applications may be customized to conform quite closely to a given shipper's operation. These client applications call upon the necessary rate servers, as needed, for the appropriate shipping rates and shipping requirements of the selected carrier.

The multitasking operating system and the client/server architecture of the preferred embodiment comprises the logical structure of the logistics management system. The physical structure, i.e., how many computers are used and how those computers are interconnected, can vary widely and still implement the above-described logical client/server architecture. More specifically, the preferred embodiment is constructed to support distributed applications in which different pieces of the total client/server structure are run on multiple machines interconnected together via a network. In general, there is no limitation on how the respective client/server components are to be distributed across the network. Thus, for example, in the system illustrated in FIG. 1, any of the computer terminals associated with stations 22, 24 and 26, as well as computer 38, may host one or more client applications and one or more server applications.

The present invention uses a plurality of program building blocks or program objects, each having a specific function within the overall client/server architecture. The building blocks or program objects which make up the presently preferred embodiment are illustrated collectively in FIG. 2. FIG. 2 is illustrated in the form of a window or folder 50 containing a plurality of icons, each representing one of the program objects which make up the preferred embodiment. A brief description of each of these objects is presented below. Further details of the manner in which these objects communicate with one another and further details of the objects construction will be presented thereafter.

The Presently Preferred Program Objects

The presently preferred program objects are described in Table I below. Broadly speaking, these objects can be classified as being client objects or server objects. For example, the objects bearing the designation "server" or "manager" function as server objects. The objects designated as "client" or "administration" (admin) function as client objects.

More specifically, servers such as rate servers encode the knowledge required to answer questions such as how to calculate shipment rates or how to band shipments. Thus, rate servers provide the knowledge regarding a specific carrier's requirements. Typically, rate servers are provided with specific details regarding a given shipment's weight or the required delivery date by a client application. Also typically, rate servers do not have user interface screens. Servers simply appear as icons on the user's desktop and wait to be asked a question by a client. Provided the question includes the right details, the server will then return the correct answer to the client. Servers can reside anywhere on a network, so they may not necessarily be visible as icons on a particular user's computer screen.

Clients are principally responsible for asking specific questions of the servers. Clients have responsibility for gathering and displaying information. As such, clients usually have user interface screens through which a user can enter data or input data through an attached scanning device.

Manager objects are principally responsible for managing aspects of the logistics management system, such as the communication between clients and servers, or communication with printers, scanners and the like. Administration objects are principally responsible for providing a user interface mechanism whereby the user may edit system settings and scripts.

In an actual implementation of the system, one supervisory server and at least one supervisory manager would be provided. Specifically, one (and only one) supervisory server is provided for the entire network. In addition, one supervisory manager is provided for each CPU that will be running client or server applications on the network. This is illustrated in FIG. 3A, which depicts a distributed architecture employing three networked CPUs 102, 104 and 106. As illustrated, the supervisory server and one supervisory manager are running on CPU 104. Supervisory managers are also running on CPUs 102 and 106. For illustration purposes a UPS rate server object and a UPS rate administration object are also running on CPU 102; whereas a shipments client and a device manager are running on CPU 106. Thus, FIG. 3A represents one possible configuration assembled from the collection of presently preferred program objects.

Although multiple CPU, distributed architecture installations represent a powerful configuration, it is possible to implement the invention on a single, standalone, CPU. This is illustrated in FIG. 3B, in which a standalone CPU is indicated at 108. As illustrated, this simple configuration includes a supervisory server and supervisory manager, along with any other program objects which may be required for the task being performed. Thus, in the example illustrated in FIG. 3B several rate servers and a packages client have been illustrated.

Referring to FIGS. 4A-4L, the presently preferred user interface will now be described. In general, the client/server program objects, illustrated as icons in FIG. 2, may each have a user interface in the form of a screen or window through which the user can enter information, make command selections, and look up information. In practice, the user interface may vary in appearance and function, as dictated by the particular task to be performed. Thus the user interface screens illustrated in FIGS. 4A-4L are intended merely as examples in accordance with the presently preferred embodiment.

Overall Icon View--System Folder

The presently preferred embodiment places icons for all user-selectable program objects in the Logistics Management System folder or window, shown in FIG. 2. In this regard, the window shows a sample set of icons for the presently preferred system. Each icon is a pointer to a seperate client, manager or server object. By double-clicking on the appropriate icon, its object is started. For example, double-clicking on the Shipments Client starts that program object running. Typically, once an object is started, it continues to run and is able to communicate with other program objects via the interprocess communications (IPC) mechanism.

Shipments Client

Shown in FIG. 4A, the Shipments client accepts user input for the routing, rating and documentation of a group of packages comprising a shipment. Multiple shipper accounts are allowed and the desired account may by selected from the Shipper "drop-box." Similarly, the service is selected from the Service box. Alternatively, the service may be set to Best Way and the system will choose the least cost carrier which meets the transit time requirements indicated in the commitment field.

The operator types or scans the Reference # (such as order #, pick ticket #, . . . ) and the system may be set to look up the associated information from one or more local and remote sources such as databases and mainframe or minicomputer terminal sessions. The upper left quadrant of the screen is to record information for the shipment as a whole.

The lower left quadrant is used to record specific information for each package in the shipment. In all client data entry screens there are several special data entry provisions. Any field which has an ellipsis (. . . ) at its right edge has additional related fields of data available to be "examined" or edited by touching the F10 key or clicking the Examine icon. A popup window with the associated fields is displayed for the user. In addition, most fields may be set up to "browse" available valid entries. They may browse from database records or from "hard-coded" values in scripts.

The Shipments client and most other clients are capable of processing shipments of mixed modes; e.g. small parcel ground, small parcel air, LTL motor freight, air freight, and TL motor freight.

Packages Client

Shown in FIG. 4B, the Packages client is designed to facilitate the entry and processing of shipments which typically consist of single small packages. Although, like most other clients, it will handle multiple modes of shipment, it is best suited for single piece shipments. If a multi-piece shipment is encountered, the user may touch CTRL-M or click the Multi button and the shipment is accomodated.

Script Administration

The Script Administration object, shown in FIG. 4C, allows the creation and editing of scripts for the modification of default behavior of the clients. A script may be triggered in various ways, such as upon the changing of the contents of virtually any field, upon the pressing or clicking of the function buttons on the client screen, upon the opening or closing of the client, and so on. In the illustration of FIG. 4C, the Shipments client is being operated upon to modify the Commitment terms based on the type of Service being used by the shipper.

UPS Rate Adjustments

Referring to FIG. 4D, the UPS Rate Adjustments program object and substantially similar objects for each of the carrier rate servers installed on the system, allow the user to adjust the discounts and incentive programs extended to the shipper by the carrier. Existing discounts may be edited, or new incentive programs not yet envisioned by the carrier may typically be created by the user within the flexible structure of this client type. Adjustments may be qualified by destination (either zone, postal code or destination country) and by weight range. Adjustments may be calculated as percentages or fixed amounts and include or exclude special service fees. If desired multiple adjustments may be created and put into effect.

Serial Port 2 Configuration

The Serial Port 2 Configuaration screen, shown in FIG. 4E, is a part of the Device Manager object. It allows the adjustment of basic serial port setup values. By clicking on the "Defaults" button, the default settings appropriate to the specific attached device are automatically entered. The "Test" button provides a facility for sending data to and receiving data from the attached device. Additional aspects of the Device Manager object are discussed in connection with FIGS. 4G-4L, below.

Document Administration

The Document Administration object, shown in FIG. 4F, allows the user to adjust settings for the UCC-128 standards of serialized container marking and Electronic Data Interchange. It allows the user to load and store consignee names and addresses and UCC-128 specific values. It also allows the user to edit the database and settings as changes may occur.

Document Server

The document server, illustrated as one of the icons in FIG. 2, allows any client to print almost any type of document, including shipping labels, waybills and manifests. The information needed to print a document is processed through a script. This allows the data to be brought in from any number of sources such as databases, mainframes, files or user-entered information from a client application. Also, one document format can serve several client applications and any particular processing needs a user might have. This is quite readily accomplished due to the fact that all data is passed through scripts. Thus, for example, one user might look up the export information for a FedEx international document from a database, while another user might wish to hand enter this data. Both of these tasks can be accomplished by a single document format and without the need for custom programming.

Device Manager Object

The Device Manager provides a device-independent means of interfacing with peripheral devices. Device drivers can be added or removed without modifying the software. The Device Manager also provides integrated testing tools.

The Device Manager of the presently preferred embodiment has the ability to monitor power, through an uninterruptible power supply connected to the system. If power fails, the other applications are notified, and an orderly shutdown of the system will take place. This prevents loss or corruption of data by sudden power outage or by subsequent failure of battery power, once the reserves of the uninterruptible power supply have been depleted. The monitoring service provided by the Device Manager can shut down multiple machines on a network connected through one uninterruptible power supply.

Referring to FIGS. 4G-4L, additional capabilities of the Device Manager object are illustrated. In FIG. 4G the communications port COM2 is shown as assigned or connected to an electronic scale, in this case a scale designated Toledo 8213 Scale. The parallel port LPT1 is shown as being assigned or connected to a printer, in this case a printer designated IBM 2380 Printer. Beneath the printer designation there appears a notation "No stock" which indicates that no particular paper stock has yet been designated.

Double-clicking on the COM2 icon brings up the configuration screen shown in FIG. 4E. Selecting the purl-down menu designated "Device" brings up the device selection menu shown in FIG. 4H. Using the device selection menu the Toledo 8213 Scale may be selected, as illustrated. Thereafter, by returning to the device manager menu of FIG. 4G, the Toledo 8213 Scale icon may be double-clicked to bring up the device settings menu illustrated in FIG. 4I.

Alternately, from the device manager menu of FIG. 4G, the parallel port LPT1 may be selected (by single-clicking) and configured by again accessing the "Device" menu option. In this case, referring to FIG. 4J, the IBM 2380 printer may be selected, as illustrated. Thereafter, the user may click "OK" to return to the device manager screen, FIG. 4G. From the device manager screen the user can double-click on the printer icon to bring up the appropriate device settings menu (FIG. 4K) for that device. Referring to FIG. 4K, note the Current Stock states "No stock loaded." If desired, the user can change the current stock by clicking on the change button. This action brings up the stock selection menu of FIG. 4L.

The client/server architecture utilized by the logistics management system affords a great deal of flexibility. The client and server program objects are designed to work independently of one another, communicating with one another through a tokenized message handling mechanism discussed below. In this way, rate server data and user data are separated from one another. The advantage of this is that when a given carrier changes the way rates are handled, the affected rate server can be modified (to change the type or amount of data stored in that rate server, for example) without affecting the user's data in any way. This separation is important since carrier requirements and carrier service options may change at any time.

The presently preferred embodiment implements an internal version numbering scheme which provides a mechanism to allow client applications to determine what version of a server they are communicating with. In so doing, the client applications are able to make any necessary adjustments or to disconnect from that server if incompatibilities are found. This provides greater reliability, since server applications are given the ability to handle communications with servers intelligently.

Separation between client and server objects also makes possible an automatic updating capability whereby a user's existing setup information is automatically merged into a new installation when an application is updated or reinstalled. This reduces the amount of setup time due to reinstallation.

Each server of the presently preferred embodiment has built-in debugging capabilities which allow server transactions to be displayed on the screen or logged to a file for later analysis. The presently preferred rate servers optimize shipments to minimize cost through the use of shipment pricing rules supplied by the carrier. Optimization occurs "live" as a shipment is being processed, or, alternatively, at the end of the day when all packages shipped are then optimized. The rate servers can directly access carrier-supplied data, such as Federal Express routing and rate file data. This allows the user to load in new rate information as soon as such information is provided by the carriers. In addition, the rate servers of the presently preferred embodiment have the ability to rate and process a package, but to withhold it from the manifest until notified to do so. This allows "pack and hold" or future shipping. This is important in an automated system where a package may be processed upstream, but not placed on the manifest until it reaches the shipping dock.

In the presently preferred embodiment client applications have editable bar code templates, to allow data to be entered via a barcode scanner. Because the templates are editable, they can be added to, deleted from or modified by the user without the need for additional programming. Clients can also process in "batch" mode, reading data from a source and sending that data to the appropriate rate servers for processing. The sources for data can include databases, files or direct connection to a host computer via serial or network connection.

As stated above, the presently preferred embodiment uses a multitasking operating system which supports the processing of multiple processes or multiple threads effectively simultaneously. This allows client applications and server applications to operate effectively concurrently. To illustrate how multiple threads operate in the presently preferred embodiment refer to FIG. 5. In FIG. 5, the left-most box designated main thread 120 represents the starting point. Main thread 120 is launched by the operating system. One of the functions of main thread 120 is to perform or control the performance of the operating system routines illustrated in FIG. 5 by loop 122. When a server is started under the operating system the main thread launches a server thread, as represented by launch thread arrow 124. The server thread is represented by box 126. In the preferred embodiment the server thread performs the necessary initialization routines 128 and then enters a loop or state 130 where it waits for a client to initiate a connection request. When such a request is received the server thread launches a client thread, depicted in FIG. 5 by launch thread arrow 132. The client thread is depicted by block 134 in FIG. 5. It is responsible for handling the client's request. Once the client thread is launched by the server thread, the server thread returns to its wait for client loop or state 130, whereby the server thread is then able to launch additional client threads upon demand. Thus in the preferred embodiment, each server that is running will include a main thread and a server thread. Client threads are generated on an as needed basis. There is one client thread for each currently connected client.

Communication between client and server, whereby requests are passed to the server and responses passed back to the clients, may be accomplished in a variety of different ways. In general, the methods of communicating between multiple processes running on the same CPU include shared memory, semaphores, pipes, queues and signals. The methods of communicating between multiple processes running on different CPUs (distributed architecture) include mechanisms such as NETBIOS, named pipes, sockets and mail slots. Collectively all of these methods of communicating between multiple processes form part of the interprocess communication (IPC) mechanism. Not all multitasking operating systems provide each of these IPC mechanisms. The presently preferred embodiment runs under the OS/2 operating system and uses LAN Server to provide operating system support for a distributed application architecture over a network. The presently preferred embodiment uses named pipes as the IPC mechanism.

Named pipes allow one process to communicate with another process as follows. The client process first requests a named pipe connection to a server process. Once this connection is made, there is little distinction between the client process and the server process, since either can communicate with the other. One advantage of using the named pipe IPC mechanism is that the client process machine does not have to be running under the same operating system as the server process. All that is required is that the operating system platform on which the client process is running will support named pipes. Thus, a client process running under MS-DOS, or under MS-DOS with windows, for example, could establish communication with a server process running under the presently preferred OS/2 system, provided named pipe communication is supported. Essentially, under a named pipe communications scheme, the server process is known by a pipe name. Programs wishing to connect to that server will use the pipe name, in a fashion similar to using a file name. For example, pipe names may take the form:

Although the named pipe IPC mechanism permits communication between any server and any client, the presently preferred configuration constrains communication to a tree-structured communications scheme illustrated in FIG. 6. Referring to FIG. 6, the single supervisory server communicates directly with only supervisory managers. In FIG. 6 the lines of communication are designated by the letter "C" at one end and the letter "S" at the other end. These indicate, for a given line of communication, which object is the client and which object is the server. Thus each of the supervisory managers communicates as the client with the supervisory server. In a similar fashion, the rate server, label server, external processing manager and device manager all communicate as clients with the supervisory managers, as servers. Likewise, the rate server administration object, the shipping client and the device manager administration object all communicate as clients with one or more of the objects appearing above them in FIG. 6 (namely rate server, label server, external processing manager and device manager). Thus it will be seen that certain objects function in a dual capacity, serving as client objects in some instances and serving as server objects in other instances.

The presently preferred communications scheme permits objects to communicate with devices, external databases and other computer systems which are not part of the client/server logistics management system. To accommodate this certain objects are given the ability to communicate with the outside world. These objects include the device manager, which communicates with hardware devices, such as printers, bar code scanners, modems, postal scales and the like. Communication with external databases and other programs which do not form a part of the client/server logistics management system (e.g., accounting software packages, spreadsheet programs, operating system utilities and the like) are communicated with through the external processing manager.

Because the presently preferred embodiment can be implemented in multiple CPU environments (distributed architecture), an announcement mechanism is provided in order to extend the communications scheme to multiple CPUs across a network. The supervisory managers are preprogrammed with the ability to send "announcements" across the network operating system according to the named pipe protocol which has been implemented. Thus, as illustrated in FIG. 6, each of the supervisory managers includes a communications pathway designated "announcements" through which communications with parts of the system operated by different CPUs are sent and received. The supervisory managers are not, themselves, responsible for coordinating this message passing between objects controlled by different CPUs. That function is reserved for the supervisory server, since the supervisory server occupies the unique position of controlling the registration process whereby client and server applications are made aware of each other in order to communicate. However, in order to distribute the announcement function, the supervisory managers are responsible for providing the actual announcement communications. In this regard, the supervisory managers act as servers, with the supervisory server in this instance, acting as the client. To illustrate this in FIG. 6, each of the supervisory managers is connected to the supervisory server by a second line of communication designed "receiver." It will be understood that these "receiver" connections represent a "reverse" client/server connection, in which the supervisory server acts as a client in order to request announcement services from the supervisory managers, acting as announcement servers.

In summary, FIG. 6 illustrates the preferred IPC connections over which client/server communications take place. The presently preferred embodiment uses a tokenized message passing scheme in which all data is passed back and forth between client and server as ordered pairs of tokens and associated data values. The purpose of the token is to uniquely identify the data associated with it. Referring to Table II in the accompanying Appendix, a sample listing of tokens is presented. Reading the columns of Table II from left to right, the left-most column lists the token name; the column to its immediate right gives a brief description of the nature of the value associated with the token; the next right-most column lists the type declaration of the value; and the right-most column gives the maximum length of the value, where applicable.

The tokenized message handling scheme is very flexible, in that new tokens can be added at any time to accommodate new features, without the need to rewrite all client and server data structures. This is in contrast to the conventional fixed field data structure used in conventional data communication schemes. In the conventional, fixed field data structure scheme adding a new data value often requires all program modules to be rewritten to accommodate the added data field. The tokenized message passing scheme of the present invention avoids this problem. If a new data field is required, to support a new feature, for example, a new token is created and only those objects which make use of the new value will need to be reprogrammed to scan for the newly added token. All remaining objects simply ignore tokens which are undefined for them.

The presently preferred tokenized message passing scheme implements automatic data type conversion. As set forth in Table II, the type definitions of all values associated with tokens are predefined, thus the tokenized message passing scheme has advance knowledge of the data types of all values. This allows the tokenized message handling scheme to perform all data type conversions, removing the need for client and server objects to perform type conversions. In other words, a server can pass a long word value to a client which is expecting a string value. The tokenized message handling scheme performs the data conversion automatically so that the server does not need to be aware of the client's data type requirements and the client does not need to be aware of the server's data type requirements.

External Processing Manager

In order to allow the client server logistics management system to communicate with the outside world, e.g. with external data bases or other application programs, the external processing manager is provided. The external processing manager is, itself, a client of a supervisory manager, as illustrated in FIG. 6. The external processing manager, in turn, operates as a server to provide external processing functions to other clients. In FIG. 6 the shipping client uses the external processing manager for this purpose.

In the preferred embodiment, the external processing manager interfaces with the REXX command interpreter supplied with the OS/2 operating system. The external processing manager is designed to receive its instructions from an ASCII file called a script file. If desired the script file can be encrypted to prevent unauthorized access. The script file comprises a list of program commands or instructions written in the REXX language. If desired, the REXX language command set can be extended to add additional commands. This may be done by embedding the additional commands in the external processing manager. The external processing manager would envoke the REXX interpreter and register itself as the source of the additional commands. In this way, the REXX interpreter would automatically pass control to the external processing manager to handle the additional commands. The external processing manager is designed to communicate directly with the REXX interpreter, passing the script file commands to the REXX interpreter and requesting the output of the REXX interpreter to be directed back to the external processing manager, where appropriate. In this way, the external processing manager is given access to the operating system and to all other application programs running on the operating system. This is a very powerful command which allows the logistics management system to interface with other applications which may not necessarily be designed to integrate directly with the logistics management system. For example, the external processing manager could use the REXX interpreter to send SQL queries to a database, in order to upload information about a customer's account from the accounting system software.

The external processing manager's ability to handle scripts provides another powerful feature. Scripts may be written for the purpose of modifying the performance of other program objects comprising the logistics management system. Scripts can be used, for example, change the shipments client to provide a reminder message to the user in the event the user attempts to ship a package without first entering the weight of the package. Similarly, a script could be written to change the shipments client to supply a convenient list of package dimension sizes, allowing the operator to quickly fill in the size of a package in the appropriate field by simply selecting it from a list. In general, the scripting language can be used to provide virtually any custom tailoring that a shipper might want to implement. The REXX language is straightforward and easy to use, thus most customizing to meet the user's requirements can be done in the field, without the need to access or modify the underlying program source code.

In running a script to modify the performance of a program object, the external processing manager uses the existing IPC mechanism. Through this mechanism the external processing manager can request a data item stored by the object or it may supply a data value as an input to the data object. In addition, the external processing manager can request that one or more operations defined for that object to be performed. Thus the external processing manager serves as an alternate to the normal user interface as a means for manipulating data or performing operations.

From the foregoing it will be seen that the present invention provides a logistics management system comprising a plurality of building blocks or client/server objects. These objects can be configured to communicate in a variety of different ways to accommodate virtually any transportation-related logistics application. Thanks to the client/server architecture and the support for distributed architectures, the overall logistics management task is readily subdivided into highly self contained functional units. When changes in a given function are required, only the object providing the function normally needs to be changed. The client/server objects provide further flexibility through the external processing manager and scripting language, whereby individual objects can be modified to provide special features quite readily on an as needed basis.

While the invention has been described in its presently preferred form, it will be understood that the principles of the invention may be extended to a wide variety of different forms. Accordingly, the preferred embodiment described herein should be considered as exemplary of the principles of the invention and not as a limitation of the scope of the claims.