sitemap home
logo Background
Crux of The Problem
The Solution

Redefining the MES
Architecture
Goals
Job Opportunities
Bibliography
D-Factory

D-Factory is a new concept in fielding
a Manufacturing Execution System.


Unlike most MES applications,
D-Factory targets small manufacturers.
Read onward to see why and how.

The MESA International Definition:
"Manufacturing Execution Systems (MES) deliver information that enables the optimization of production activities from order launch to finished goods. Using current and accurate data, MES guides, initiates, responds to, and reports on plant activities as they occur. The resulting rapid response to changing conditions, coupled with a focus on reducing non value-added activities, drives effective plant operations and processes. MES improves the return on operational assets as well as on-time delivery, inventory turns, gross margin, and cash flow performance. MES provides mission-critical information about production activities across the enterprise and supply chain via bi-directional communications."

 
 Background
Manufacturing Execution System (MES) applications provide real-time visibility and control of plant floor operations.
MESs occupy a special niche between those applications traditionally associated with the factory floor:

     dotplant-floor automation
     dotsupervisory control and data acquisition systems (SCADA)
and those applications associated with the front office:
     dotenterprise resource planning (ERP)
     dotsupply chain management (SCM).
     dotcustomer relationship management (CRM).

MESs excel at batch-oriented manufacturing operations (e.g. build-to-order, box-build assembly, printed circuit board assembly, automotive assembly lines). They are not well suited to continuous process environments (e.g. petrochemical refineries).

MESs are used to tighten the integration of factory-oriented shop-floor operations with customer-oriented front-office applications. The intention is to automate the data interchange between the two systems. It makes that interchange more efficient and less error-prone. It can allow customers direct access to shop-floor operations.

Traditional MESs have been relatively unsophisticated client/server applications targeted at medium- to large-sized manufacturers with mid- to high-volume operations. The client application is typically MS/Windows based. The MES server application usually requires a medium- to large-scale Unix computing platform (Solaris, HPux). Additionally, an MES installation requires an RDBMS (e.g. Oracle, Informix, Sybase) system, also usually running on a medium- to large-scale Unix system.

Obviously there is a significant amount of infrastructure, both human and hardware, required to support such applications. Obviously there is an associated cost. In addition, the MES application itself isn't cheap: the customer is billed for delivering, installing and configuring the MES product itself, plus an additional per-seat charge (meaning, the fee for a single client workstation for one operator). There may be an additional charge to install a new RDBMS, or at least to configure a new instance in an existing one. The total cost is hefty enough that there can be a tiered pricing structure with stepped discounts for large numbers of clients. And there may be additional charges for customer service, long-term maintenance and access to future upgrades and patches. Not that this is unfair: MES suppliers are certainly entitled to make a profit for their work, as long as there is sufficient competition between suppliers to keep prices "reasonable" in a particular market segment.

Recent wrinkles in MES technology are driven by the ongoing shift of customer computing environments from specialized MS/Windows client applications directly connected with large database server applications, to simpler web-based clients and a distributed server architecture. This trend follows the general shift of customer applications from simple client/server applications to the vastly more flexible multi-tier Java Two / Enterprise Java Beans (J2EE/EJB) web-based architectures with light browser-based clients.

This new architecture has several big advantages over earlier client/server technologies. First, it scales nicely for larger transaction loads and bigger page hit rates. Second, it is more secure since it isolates the sensitive, big-ticket database installations behind a controlled firewall. All database activity is buffered from direct contact by potentially-nasty client applications.

This new architecture also allows the client-side use of a variety of new thin-client technologies to interconnect with multiple database services via a set of specialized mid-tier application services. There is new opportunity to monitor or inject data in the specialized data interchanges between the mid-tier servers. It opens new possibilities to combine and manipulate data sources and redirect data flows.

  The Crux of the Problem
Nowadays, to stay profitable a manufacturer's day-to-day business must stay synchronized in near real-time to customer demand. Big manufacturers have deep pockets. They can ride through market fluctuations such as temporary supplier outages or a drop in product demand without much adjustment. But small manufacturers don't have the luxury of that money cushion. In fact, they live or die based on their access to cost effective and lean manufacturing techniques. They don't have the infrastructure or the cash either to maintain stocks of incoming supplies or to warehouse finished products if their processes aren't tuned well or don't give a decent yield.

In fact, what these small manufacturers really need is a scaled-down, cheaper version of that traditional MES.

  The Solution
We believe the small-scale manufacturer, possibly even as small as mom-and-pop operations, is an ideal market niche for that simpler and cheaper MES. There is no product currently targeted at that market, and economies of scale will probably prevent existing MES suppliers from ever reworking their products to reach that market. We want even the little guys to afford an MES.

The goal of D-Factory is to re-invent the MES. But not in a vacuum. We want to build on previous work, to keep the characteristics of big MESs that are still useful in a scaled-down environment, and to eliminate the ones that are not.

We wish the software components of this new MES to be freely available to developers and to users, so we propose to release all work under the GPL.

This project is in its startup phase.

If you would like to participate in setting up requirements,
please have a look at the posted job opportunities for this project.

Many thanks to SourceForge for their most excellent services! SourceForge Logo