MCA Solutions - A Strategic Service Parts Management Platform
MCA Solutions, a Philadelphia, PA company, is not only one of the few companies I know of that has an advanced strategic service parts management solution, but one of the very few that only does service parts management. Recognizing that many large manufacturing, semiconductor, high-tech, aerospace and defense companies often have tens of millions, if not hundreds of millions, of dollars tied up in inventory, and that an inventory planning and optimization solution that is off even by a few percentage points can cost these companies millions, if not tens of millions, of dollars annually, the founder of MCA Solutions, Dr. Morris Cohen, who has worked with IBM, Cisco, Applied Materials, Intel, GM, Saturn, Teradyne, and the U.S. Navy, decided to focus the company on this problem alone.
Why? Because the problem is a lot harder than you think. Just like a product has a life-cycle, so does a service part. Not only do you have to accurately forecast how many replacement parts you're going to need in your network (as well as where they need to be), you have to manage the return, repair, and re-introduction of the repaired part into your inventory. (Remember, many parts are sub-assemblies because it can be too time consuming to replace an individual part - so it needs to be repaired once it is replaced; just like your IT department doesn't throw out the desktop they just replaced when only the hard-drive needs to be replaced.)
To accurately solve the problem, MCA Solutions allows you to model your entire multi-echelon parts demand network. What does this mean? You can model all of your primary (warehouse) locations, forward locations, forward-forward locations, etc. to as many levels as you need; you can define all of the production lines, aircraft, or other equipment at each location; define the required replacement parts and desired availability and / or target stock levels for each part; define any and all (performance-based) contractual commitments if you are in the business of servicing lines, aircraft, or other commitments for your customer; define historical demand, service requirements, or maintenance plans; and specify the best type of statistical model for the part in question (poisson, normal, or negative binomial - as low volume, high-volume, and sporadic demand parts need to be modeled differently), as well as any location or usage-specific criteria that influences demand.
Furthermore, MCA Solutions' platform not only allows you to strategically plan cost-optimal inventory levels for target stock and availability levels, but also takes into account current network stock levels and will give you an executable tactical implementation plan which will tell you what needs to be shifted between locations, what needs to be ordered - and when, and which parts should be repaired (and when) and which parts should be retired. In other words, not only does their solution understand product life-cycles, but it also allows understands the entire part life-cycle.
How well does it work? For their target industries, very well. It was chosen by the Navy, who spent almost a year exhaustively evaluating COTS (Commercial Off The Shelf) solutions against their own in-house solution, it's used by KLA-Tencor and Cisco - who have some of the most extensive parts supply chains in the IT world, and their solution has been chosen by SAP as their preferred parts planning solution. Furthermore, it's very well designed. You can work at the aggregate network, network (as it allows you to define different part networks if you have to meet different geographies, different environmental regulations, or just want to separate your internal service networks from those of your customers), forward location, location, equipment / contract, or part level, depending on your need; you can compare the current plan to various "what-if" plans that let you see how your altered stock levels / availability levels affect cost or how shifting forward locations (central warehouses) changes stock levels and affected costs; and you can do extensive reporting, graphing, and, if required, data export to Excel (and Power Point). Plus, you can export orders to your external procurement / ERP / MRP systems and import supplier response data. If the lead-times in the responses differ from what the plan expects, the system will automatically update and re-balance the plan.
If you're in one of their target industries, it's certainly worth an investigation. Not only is it designed well, but it appears to be very efficient. The average response time for an update even in a fairly sophisticated what-if network model (with hundreds of locations and thousands of parts) is under two seconds. That's impressive where optimization is involved given the complexity of a multi-echelon network model.
Very interesting post.
1) Who is the competition, and how do they stack up to MCA Solutions?
2) Basing parts replacement on a Poisson arrival model seems wrong to me (because parts age, so there have to be weighting factors etc. to account for increasing arrival rates with time). However, I freely admit I'm not a pro at this, so I'm happy to be educated.
3) Probability distribution functions are not concepts that the average MBA understands, at least not many of the MBAs that I know. How does the software allow naive users to pick the right PDF for the replacement model? Note that this question is akin to the oft-repeated complaint about optimization in general, namely that it is "too hard" for business users to master.
I was hoping MCA Solutions would chime in on this, but since they didn't, here's my first take:
(1a) Companies like Logility, New Breed Logistics, and Servigistics as well as i2 and SAP.
(1b) Depends on the vertical, the relative experience of the company, and the breadth of implementation desired. I think Servigistics stacks up well, especially in reverse logistics, for CPG. I'm not familiar with Logility and New Breed Logistics (yet - maybe they'll take the hint and contact me), and I think both MCA Solutions and Servigistics out-do the offerings of SAP and i2.
(2) The problem is, for certain parts that are rarely used and whose precise breakdown time is unpredictable, there really aren't good models. However, if you look at the problem in aggregate, applying a Poisson arrival model across the board on all parts which fall into the category will, on average, average out and give you a better solution than a model that assumes lots of parts on a regular basis will be needed.
(3) My understanding is that the system chooses "default" distributions based on part type and usage, and that they (as I) strongly recommend that it is set up by the provider in collaboration with the engineering & R&D teams. Then, all the MBA's have to do is fiddle with the "availability" and "lead-time" and "on-time" targets to run their models.
Thanks Michael for the excellent write-up on Strategic Service Management and MCA. Michael gave a good response to Eric's excellent questions, as Michael requested, I will add a response from MCA's perspective.
1) Michael's write up on the competitive environment was helpful. I will add that MCA remains focused on the service supply chain and has a long history of innovation and success in the market and is emerging as the leader. Instead of trying to do all things related to service, we chose to partner with SAP to complement and extend their solution and the partnership is paying off in several joint deals and a large pipeline to make us the default solution for any SAP customer looking for a solution.
2&3) Eric noted that a poisson distribution is not the right approach for every part. MCA has a range of forecasting techniques and probability distributions and the appropriate one is selected automatically based on the attributes of the demand data associated with the part, with available planner overrides as desired.
Rather than requiring a sophisticated user, the intelligence in the software means that the planner does not have to have a deep understanding of the underlying algorithms. Instead he/she can focus on managing exceptions and working the plan rather than spending time determining the best approach for forecasting each part.
Anyone who wants some more background on our approach can send an e-mail to andreae@mcasolutions.com.
Thanks, and keep up the great coverage.
tim
There are indeed other options besides the companies mentioned in this article. Baxter Planning Systems has been focused on Service Parts management since 1993. Baxter clients include Kodak, Fujitsu, Network Appliance, Agilent, Alcatel-Lucent, Johnson & Johnson, Philips Medical, Atlas Air and many others.
Baxter’s modular planning solution, Prophet by Baxter uses a unique total-minimum-cost approach to service parts planning. Rather than begin with a pre-determined service level and work backwards to find a compliment of parts to stock, Baxter balances the effective cost-of-stockout with the costs associated with carrying inventory. It’s through this balancing process that an optimal lowest cost target stocking level (and service level) is achieved.
End-customer install base information including service contract terms are used extensively within Baxter’s tools to drive the cost-of-stockout, network modeling, probability of part usage and ultimately, the stocking plan. The resulting analysis provides an optimal part/location field-stocking plan and also arms the client with a method to intelligently redistribute parts between locations. Baxter also works with 3PLs and client ERP systems to generate replenishment and redeployment orders.
Prophet by Baxter is a highly scalable and automated Software as a Service (SaaS) planning solution. Baxter also provides an optional Managed Services solution that includes professional on-demand outsourced planning services.
Mark Anderson
VP, Supply Chain Solutions
Baxter Planning Systems
manderson@bybaxter.com
Eric-
In reference to your first question, Servigistics is the Service Parts Management market leader, with over 100 clients across a variety of market verticals, including high tech, aerospace, motor vehicles, telecommunications, medical equipment, consumer durables, and heavy industrial.
Our clients include Blue Chip companies like IBM, Sun, AGCO, NCR, Embraer, ATR, Rolls Royce, Dell, CCE and Sears. In addition, our solution offering isn't limited to Service Parts Management - it includes service knowledge, service parts pricing and service workforce management - this is what the market calls Strategic Service Management.
As for your questions in 2 and 3, feel free to email me and I'll have our people respond in-depth.
srentner@servigistics.com
Thanks for your interest in this space!