|
Newsletter |
To subscribe (or un-subscribe) to our Newsletter,
just click on the link below.
Subscribe |
|
|
Smart Software |
|
|
|
| Manufacturing's magic pill? |
|
Written by Peter Fretty
|
While we all know there is no “magic pill” that can solve all the
problems associated with connecting the shop floor to the top floor,
true connectivity might be closer than you think.
When searching for software systems, manufacturers carry around a large catalogue of requirements as well as the perception that they can’t ALL possibility be met.
While they are smart enough to realize that there is no one solution, some lag behind in implementing some of the software that is available, and need to grasp what these tools can bring to their companies, which include:
- Better tracking of key performance indicators;
- Real-time monitoring of system and production performance;
- Real-time visibility into supplier performance and product availability;
- Simulation of production and plant processes to avoid bottlenecks, constraints;
- What-if scenarios to help with production planning and scheduling;
- Event exception alarms and sensors that trigger actions when performance parameters or machines get out of specification; and
- Predictive maintenance capacity to avoid unscheduled downtime.
Understandably, while there is no one complete software package that can comprehensively meet every item on this wish list, progress is being made.
Analytical view
According to Alison Smith, senior research analyst with Boston, Mass.-based AMR Research Inc., the problem is that even though manufacturers know they need to do a better job embracing software technologies, it is usually unclear where the actual problems exist.
“There is an awful lot of technology out there that can drive operational excellence, but the fact of the matter is that we have neglected spending money on our assets outside automation,” says Smith.
There is no denying that even with extensive au
tomation investments, the marketplace still is overcome with legacy assets, but this does not mean that it is impossible to capitalize on what current software systems have to offer. Instead, Smith suggests that it points to the glaring need to buy and implement more sensors that can instrument what is already in place.
However, for this to work, Smith says companies need to look ahead and do away with the “break to fix” model. “This is where manufacturers would do well to have handhelds complement the latest software to pick up information from all of its assets. With this type of capturing in place, you start to understand the performance information.”
Once manufacturers understand these factors they can effectively bring the “what if” scenarios into play. “The first step is to bring the process into control, so that it is possible to understand the variable before you start playing around with the data,” says Smith. “This is the only way to effectively understand how to downplay constraints and do well with your strengths, but you need to recognize your problem areas in order to find the solutions.”
James Keane-Murphy of Westchester, Ill.-based itelligence Inc. adds that throughout this process manufacturers should attempt to rid themselves of what he calls, “the relentless pursuit of the guilty.” “Sure the traceability is what gives you the visibility as well as production support that allows accurate decision making, but the system cannot be a dictator,” he says. “Instead it should present proposed routes that you can either accept or reject. Done right, these are what identify bottleneck constraints that can ruin production schedules.”
Warning signs
The first step in selecting the most appropriate smart software is looking for the warning signs pointing to shortfalls in existing systems. While some signs may jump out, it is those that go unnoticed that can truly wreak havoc on the bottom line.
The transactional process versus real-time is one test for prospective software to pass, says Matt Tormollen, CMO with Irvine, Tex.-based Pavilion Technologies. “Does software provide the appropriate contact between the two levels of data?” he asks. “Being able to provide real-time data as it relates to a transaction is a basic requirement for success.
“For instance, what comes to the plant level may include a lot of characteristics that need to be identified and packaged so that the order is not bogged down by the time it gets back to the ERP top level. Even looking at performance measurement reporting if you define at one line, one site, one plant, you need to make sure that the information relates accurately.”
Kevin Dwyer, president and CEO of Oakville, Ont.-based Shoplogix, agrees, adding that the when a company embraces poor collection techniques it usually results in discrepancies between recorded production information and actual production numbers.
“How a business handles this process (manually re-keying or embracing sensors combined with powerful software tools) will unquestionably impact the accuracy of the data used for decision making,” he says. “The attention a business pays to this aspect ultimately impacts the financial well-being of the business.”
When manufacturers experience problems in meeting customer demands it can signal a failure of the company’s software strategy.
According to Andre Pino, vice president and CMO of Marlton, N.J.-based Acsis Inc., this often comes to light when RFID and compliance labeling requirements mean implementing a separate point solution. “Capturing data and effectively using it for the business should not be an inefficient process that carries additional costs,” he says. “Instead, you need to be able to respond to supply chain problems in a timely fashion and avoid receiving complaints that you can’t tell a customer the status of their order.”
Pino adds that when the solution is adequate, it should be relatively easy to make improvements to the floor processes and business owners will have a solid understanding of order flow throughout the organization. “Otherwise the cost of capturing data in manufacturing and warehousing increases and impacts profit margins, and it is quite difficult to gain additional shop floor efficiency by automating manufacturing processes.”
Businesses need to be able to pull data off of every machine that is running a job or task on the plant floor. Such machine truths are essential and can only come from devices connected directly to the machine and independent of human intervention. Understandably, most plants will have a very wide range of assets on the plant floor, but a state of the art software product should easily address this disparate or heterogeneous operating environment.
Connectivity also suggests that the data captured from each asset should be easily aggregated and used to provide a consolidated view of how well or poorly the plant is performing.
“Connectivity should also be extended to a variety of ancillary devices like PDAs, pagers, cell phones, laptops, etc.,” says Dwyer. “This automated connectivity enables proactive communications from the machines on the plant floor with a variety of third party devices. Proactive alarms and alerts regarding preventative maintenance, quality issues, production statistics and other predetermined business rules can easily be passed on to those that have a need to know.”
Making the switch
Once the decision has been made to invest in software, manufacturers need to be aware that solutions are rarely turnkey and often require careful consideration prior to committing money and time.
According to Tormollen, when upgrading, firms need to know that there is commonality in their collection and aggregation of data regardless of which software product it purchases. “This is what provides the metrics. A lot of the information only goes to ERP as needed from a reporting standpoint when needed,” he says. “You need to think about performance and then enabling querying that allows management to ask intelligent questions.”
Dave Gleditsch, CTO of Boulder, Colo.-based Pelion Systems also warns about ignoring the adoption aspect. “When people do not trust, they do not make sure that the data is fresh and consistent with what is really going on,” he says. “You need to be able to effectively connect the dots from information to the concrete improvements within the business that really materialize into anything of significance.”
Upgrading is also the time to take a macro view of whether there is more value stream as opposed to function thinking. “Can you view at a macro level to provide information on a regular basis?” asks Gleditsch. “You need to know that things are the way you need to view them, not the way something forces you to view them.
This is all about the buy-in and acceptance that results in success. Is the software flexible yet consistent? There needs to be flexibility that meets needs while still having a solid level of standards that run an operation in a positive and consistent manner. Performance indicators and all the great data is vital, but you also need to be able to look at the present and the future, not just the past to accomplish these tasks.”
Despite the promises software developers offer up, most manufacturers are smart enough to realize that no one solution will ever be the elusive cure-all for its production woes. Yet, at the same time, it is important to realize that the industry has made great strides, and that with some impressive new capabilities they may at the very least alleviate some of the pain. nanm
Peter Fretty is a freelance writer and a regular contributor to Advanced Manufacturing. You can reach him at:
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it
|
|
Archive 1999-2006 |
|
View older Site files below
|
|
|
|
