I know a baker who is technologically astute.  He runs his bakery on JD Edwards EnterpriseOne ERP software, because EnterpriseOne features Advanced Pricing, which can handle sweet discounts on doughnuts and tiered pricing on cakes.

Using the EnterpriseOne manual, my friend the baker has figured out how to properly set up his pricing hierarchies and schedules.  He has also figured out how to price a baker's dozen of doughnuts. He simply sets up a list price and a discount pricing adjustment with Basis Code "5" (add on amount), Factor Value Unit of Measure "BDZ" (13 EA [each] == 1 BDZ [baker's dozen]), and Price Partials = "0" (no).  The discount is equal to the price of one doughnut, so that each time someone orders thirteen doughnuts, he receives the discount and pays only for twelve.  Similarly, if he orders 26 doughnuts, the thirteenth and twenty-sixth are free.  The price partials flag makes order entry simple because EnterpriseOne is smart enough to know that the customer gets the discount only after reaching the whole baker's dozen.

Now, my friend the baker has a challenge.  (more…)

Last time, we discussed a method whereby we could report on the efficiency of a crew by using indirect labor codes, and comparing the total labor on a production line to the total estimated labor for jobs produced on that line.

As we mentioned, we are unable to get a precise breakdown of labor hours by job, but we can get a fair estimate by using the weighted allocation that Epicor uses.  When an employee clocks directly onto multiple operations at once, Epicor will allocate hours to each operation based on the weight of that operation's estimated hours.  Let us make up an example: (more…)

J shared with us this crazy error message, which demonstrates that it is possible to crash the system so hard that it cannot even finish writing its own error message.  It just gives up, grabs the rod and reel, and walks out.

System Crash
System Crash

 

Let us say that your company has a production line style operation where it would be inefficient for operators to leave the line to clock onto and off of jobs.  Managers would like to use labor collection to analyze efficiency and profitability, but there are enough jobs that the time spent clocking on and off of each would kill efficiency and be cost prohibitive.  It is not worth the effort.   At first you thought Epicor could handle some type of "crew punching" where a team leader could clock onto a job and automatically his crew would be clocked on at the same time, but you found that this is not the case. (more…)

Every day at Deft Flux, we are mapping unique paths from problem to solution, and we like to share them with you from time to time.  Today, our question is a common one -- "How can I automatically run and e-mail a Crystal Report?"

Let's consider this case where you want to schedule a report to be run and delivered to some users via e-mail.  For our readers who use Epicor, scheduling a report is as simple as setting up the system agent, loading the report, selecting the schedule, and pressing submit.  For the e-mailing, Epicor offers a module called Advanced Print Management (APM) and it would be the simplest way to e-mail reports in Epicor.

It may be, however, that your company does not have a license for APM or you may need to report on data outside of Epicor -- perhaps another relational database or an XML file.  (more…)

During an upgrade to JDE EnterpriseOne you will need to list any changes that have been made to your data dictionary, and using SQL or even a UBE, most developers would have no problem coming up with a way to compare the pristine data dictionary with the latest version.  We thought, however, it might be nice to give those with less experience in this line a jump start.  So here is some SQL that could be used to detect new data dictionary items and changed alpha descriptions.   (more…)

Epicor has been a substantial part of our work for the last 18 months.  One client has made a wise decision to not customize the application, so most of the time, we have been working with business processes to get the most from the software and make people's jobs easier.  There have been, however, the inevitable situations where an opportunity exists for someone to make a mistake and a simple tweak to the software will totally eliminate that mistake.  For the business, it is an easy decision.  Customize.

This is where Epicor  holds an advantage.  Building a business rule into the software does not introduce the complications that a normal software customization would.  Epicor provides us with a number of ways to insert rules into the workflow of the software, and they can easily be exported, either on their own or in combination with other tweaks, using the Package Manager.   For what it gives you, this rules engine is quite powerful.  For example, we can add a rule to the Sales Order Entry form that will warn a user if a customer's purchase order number has been used before, or force the user to enter a disposition for non-conforming material.  Moreover, adding these types of rules takes a minimum of effort, say less than an hour.

We are pleased with the way Epicor has provided a way for small businesses to add minor business rules to their ERP software.

We recently had the chance to see Watson, IBM's Jeopardy-playing super-computer, mercilessly pummel Ken Jennings and Brad Rutter, Jeopardy champions.  While we believe that Ken and Brad knew just as many or more questions than Watson and Jeopardy is as much about the buzzer as the questions, Watson providing so many correct questions is nevertheless a feat.  We also got to see the Nova episode that went into more detail about the development of Watson.  IBM tried to use a rules based answer engine first, and then went to a pattern matching engine.  We immediately recalled the fellow who trained his pattern matching spam filter to play chess.  Anyway, based on what we have seen of rules based spam filters and pattern matching (Bayesian) spam filters, we are not surprised that Watson's pattern matching question engine performed much better than the rules based engine.  You may say we are surprised that IBM did not try that first.

To conclude:

  1. Let us remember that although the question algorithms give the computer the appearance of intelligence, this computer is no more intelligent than any past computer.  That is, it is just pretty good and really fast at matching patterns of a certain type.
  2. Maybe we need more information, but based on what we know, we would not say Watson is a major breakthrough.  IBM just was able to put together a sufficiently powerful computing system to store and retrieve all of the necessary information.  We give credit to the smart guys at IBM.
  3. In the end, it was a buzzer pressing contest and no one can beat a computer at that.

Builders of information systems that work as they're intended to.

WORK WITH US
linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram