jump to navigation

Half the Data You Need to Run Your Business is Outside May 23, 2007

Posted by Jeff in Business.
add a comment

We’ve had a chance to talk to many operations managers in our customer space.  What many are realizing is to perform the operations management function, much of the information is located outside their business.Flat Earth

What are some examples of this information?

  • Transportation Information
  • Currency Information
  • Competitor Actions Information
  • Reference Part Catalogs
  • News and Weather
  • Geographic Information
  • Regulatory Information

We have generated this list based on real-world input, and defined a set of exception resolution scenarios that highlight the role of external information.

How is the information used?

Here is an example:

  • Transportation information is often quoted in foreign currency.
  • Weather information can be helpful in determining if a shipping path is available.
  • A part catalog used to determine possible substitutions.  Also, the system has maintained a record of prior substitutions, to determine which of the possible substitutions have been shown to be a most effective match.
  • Geographic information and transportation information is used to find alternate carrier.
  • Competitor pricing information can help determine a price for your products.
  • Compliance information can be critical in determine of a shipment is valid under RoHS regulations.

How is Information Fetched?

Typically with web services API’s.

Examples of web services deployed today are those supported by information utilities such as FedEx, Amazon, eBay, and more.  By using the web service, you can programmatically post a listing on eBay, order a book on Amazon, or schedule a shipment on FedEx.

At a simple level, web services are a remotely invocable operation.  While this has been done several times before (RPC, DCE, CORBA, etc.), web services are a more open and standards-driven version of the concept.

The second most common case is a feed that produces RSS or XML content,  A number of weather, news, and finance sites work this way.  Another example of this are sites that provide parts catalogs, equipment databases, pricing guides, compatibility/compliance guides, etc.


Scenario for Exception Mgt and Options May 13, 2007

Posted by Jeff in Business.
add a comment

The following scenario was shown at the recent Software 2007 Conference.

In this example, we start out with the Serus application running.  It is processing fulfillment requests (orders) from a number of users, as well as getting updates about WIP and shipments in progress from the back-end sources.

As updates are processed, the system is recalculating delivery dates and comparing them against the requested delivery date.  Internally, this means that the system is using the business process model of production, fulfillment, and shipment, and generating forward estimates of dates and events using that model.

The “nominal” display for a user’s session of running the application would be a dashboard that shows a number of indicators regarding shipments, orders, etc.  The details of that dashboard design are covered in a number of other documents, so this is just a mockup.

Software 2007 Demo 01

When an order reaches a delayed shipment status that is within a defined exception notification conditions, it requires intervention on the part of the operations manager.  A notification appears, as shown below.

Software 2007 Demo 02

 Now the initiative has shifted from the computer system to the operations manager.  This “change over” is controlled by the notification and exception handling conditions:  typically the system would automatically carry out corrective processing within limits, such as small orders, non-critical orders, or when the delay is within tolerance.  Here we assume that a “major” exception has occurred.

Since the Serus application has an internal model of the steps in the manufacturing and fulfillment process, that model contains a description of a number of possible alternatives that can be considered.  It now begins to generate “what-ifs”, or options, for each of them, as shown below.

Software 2007 Demo 03

The options will be summarized in another part of the screen.  Each option has a name or description, and a quantitative impact assessment.  For instance, to deal with a transportation delay, the impact is going to be measured in terms of cost to correct the problem, and what the result is against the problem.  In this case, it will cost $400 to carry out an expedite, and that result is that the order will be completed in 2 days.

There are a number of back-end operations and data fetches going on to aid in filling in this information.  The next option shows uses more of this information, as shown below.

Software 2007 Demo 04

In this case, the system is working down the path of replacing the portion of the order which is causing the delay with an alternate part which is not as affected.  For instance, this may require the system to interact with a parts catalog vendor to determine the possible substitutions.  Also, the system has maintained a record of prior substitutions, to determine which of the possible substitutions have been shown to be a most effective match.

This score of matches based on past information is an example of the software/content ecosystem providing a guide to making a decision.  In this example, we have reduced the match to an “Ecosystem-provided” score that shows that the part coded “OPN-7400A from vendor Avnet” has been positively selected twice as often as the 8400 part, and so the system has picked the 7400 part to feed into a “what-if” scenario shown back on the options list screen, indicating that this option will only cost $200 but will take 4 days to carry out.

Clearly the operations manager would be able to override and/or reprioritize the selection of the substitute part, and other parameters of this option, but such details have not been shown here.

Next, the system moves on to another type of option, which deals with geographic and logistic information.  This is shown in the next screen.

Software 2007 Demo 05

The option in this case is built around changing the routing of the part, on the basis that that the problem isn’t due to the part not being available at the originating site when needed, but that it can’t get to the destination site when needed.

To perform this check, the system must look up geographic information and transportation information.  For instance, it might for an alternate carrier.  Running the resulting information back through the business model of activities for shipment fulfillment, the system concludes that this option will cost $300 and and will take 3 days to carry out.

Now it is up to the operations manager to select among the 3 options.  As is typically the case, there is no option that is best in both the cost and the time column!  The user makes a selection as shown below.

Software 2007 Demo 06

The “Change Routing” option is selected, and the Process Engine starts replanning.  The Alert is then taken down as it has been resolved.

What isn’t shown here is that the system should be gaining new information from the handling of this alert.  For instance, the system might generate a preference for the new routine, and apply this to future situations, in order to proactively reduce the alerts

Serus Presentation by Indu at Software 2007 May 12, 2007

Posted by Jeff in Business, Event Reporting.
add a comment

Indu Navar was recently asked to give a presentation at Software 2007 on the Serus approach to integration of content and business processes.  A video of the presentation is available here.

The following diagram shows the architecture of our solution (click for a larger view).

IOM Components

The elements include:

  • Predictive Analytics
  • Process Collaboration
  • Master Data Management

The Need for Speed May 10, 2007

Posted by Mike Lazich in Business.
1 comment so far

Today’s operations managers face an ever-increasing number of concurrent internal and external (i.e., outsourced) manufacturing activities that must be planned, executed, and tracked to carry out the complexities of modern manufacturing.  At the same time, they must be acutely attuned to the highly dynamic nature of today’s supply and demand signals.  Even as product lifecycles continue to contract, delivery requirements are becoming more demanding than ever, thus driving the “need for speed” in all aspects of operations management. (more…)