Print Posted by Blumberg Advisory Group on 07/17/2017

Innovations in Service Lifecycle Management (SLM): The Benefits of Model Based Reasoning

By: Michael R. Blumberg, CMC
Innovations in Service Lifecycle Management (SLM): The Benefits of Model Based Reasoning

White Paper - 

In 1959, during remarks at the United Negro College Fund in Indianapolis, Indiana, Senator John F. Kennedy of Massachusetts said, "In the Chinese language, the word 'crisis' is composed of two characters, one representing danger and the other, opportunity." He was speaking of the compound word "weiji" (wei ji) which since that time has become a staple meme or buzzword for American business consultants and motivational speakers, as well as gaining popularity in educational institutions, politics, and in the press. Though linguistically imprecise, as the character "ji" has many meanings, the point remains well taken, as opportunity frequently arises from a crisis, disaster, and what is often called, the moment of truth.

As the automobile became ubiquitous in the 1950s and 60s, people were dying at the rate of 50,000 + per year on America's highways. This gave rise to higher government safety standards and the development of safer cars, in part as a selling point for car dealers. Today, with three times as many drivers on the road, that death rate has been cut in half, and safety engineers, airbag manufacturers, onboard computer makers, and a host of others have been among the benefactors along with consumers.

Early warning systems, missile defenses, and 24/7 ready states of preparedness were crucial in averting nuclear annihilation during the cold war. Preparedness is essential to avoiding a crisis.

The necessity of inventing microwave dinners in order to have something to put in the newly invented microwave oven has brought convenience and efficiency to millions. Sometimes, Invention is the mother of Necessity.

Industry Trends

The truth is that every impending danger or an unforeseen development, commercial or otherwise, has within it the opportunity to change course to a new and better direction. Capital Equipment manufacturers currently face this situation more so than ever before. While it has been more than 5 years since the Global Financial Crisis rocked the world, Equipment Manufacturers see both peril and prosperity at every turn. Indeed, study after study by Economists, Accounting Firms, and Financial Analysts report a muted outlook for an economic recovery in this sector. Those who are poised to seize an opportunity and mitigate crisis will most likely prosper.

While selected segments of the High Tech Industry are experiencing growth, it is occurring at a stagnant or decelerating rate. New sales seem to come from pent up demand among customers who have waited as long as they can to purchase new technology. However, many manufacturers face intense downward pressures on price, coupled with market share erosion from lower cost competitors, while increasing labor costs are further straining corporate profits. Furthermore, political and regulatory uncertainty, and the emergence of disruptive technologies such as "cloud computing" continue to represent the largest threats facing equipment providers.

The trend toward commoditization of technology has resulted in an environment where buyers are more discerning than ever before. Not only do they expect new products to be outfitted with the latest and greatest technology, they also expect the seller to put more time and effort into educating them on why their product is the best product to purchase. In essence, buyers expect to get more for their time and money. This is not just limited to the product purchase itself but to service after the sale. Sure, in the past buyers considered the Total Cost of Ownership (TCO) in the purchasing decision. However, now decision criteria include information on the level of service availability or uptime they can expect from the product. The fact of the matter is downtime costs time and money so end-users want to hedge this risk by building uptime availability into their service contracts with the manufacturer.

In order to thrive in the current global economy, manufacturers must adopt a strategy of constant innovation. This is no longer a choice; it is a must. Innovation is driven by a desire for self-preservation and renewal. Ultimately, the most effective form of innovation is created when a company adopts one or more disruptive technologies in their products and/or services. Examples of these disruptive technologies include smart mobility, machine to machine (m2m), cloud computing, and "big data"/analytics. According to a recent study by Ernst & Young, the integration of disruptive technologies into products and services account for the differences in growth rates between those manufacturers who have experienced double digit growth versus those with double digit decline.

Challenges to Service & Support

Service & Support is not immune to the challenges of commoditization. Continuous downward pressures on service pricing combined with the requirements for higher levels of service quality have forced manufacturers to continuously seek innovations in services delivery. The deployment of disruptive technologies to service delivery processes has been one way in which manufacturers have been able to achieve this outcome.

Examples of these disruptive technologies include:

  • smart mobility to improve field service productivity
  • cloud-based software solutions to automate and optimize service operations 
  • big data & analytics to fine tune and improve service delivery activities 
  • Machine to machine technology to obtain real-time notification about impending service events

The arguments for deploying this technology are compelling and supported with impressive case studies and benchmarks with respect to the Return on Investment (ROI) and the productivity & efficiency gains achievable. Despite these reports, most if not all of the advancements are focused on the ways technology improves field service productivity. Indeed, Field Service labor is the single largest expense in managing a Service organization and service revenue is the lifeblood of the business. As such, any technology investment that results in productivity gains will have a tremendous benefit on the bottom line. Other often cited advancements are focused on the impact cloud-based solutions have on capturing missed or lost revenue opportunities, and minimizing service (e.g., warranty/contract) liabilities. At the end of the day these new technologies have been utilized primarily as an enabler for doing things faster, better, and cheaper as opposed to being a facilitator for doing new things in the realm of Service & Support.

However, facilitating technologies are found within most Service Lifecycle Management (SLM) platforms. These platforms include functionality such as Dynamic Scheduling (DS), Knowledge Management, and Service Parts Optimization (SPO). These are technologies that were once disruptive and are now considered to be commonplace. Furthermore, these focus on making field service resources both more productive and more efficient. For example, SPO results in reduced inventory investment while improving inventory availability levels. These technologies are being integrated with Disruptive Technologies to get more "bang for the buck" so to speak.

SLM platforms such as SPO and DS are focused on optimization. Optimization is about making the highest and best use of resources. In service, optimization usually involves searching multiple solutions or strategies for the best (i.e., least expensive, most accurate) solution in the shortest period of time. For example, find a strategy that results in the highest profit and the fastest response time. Optimization is all about getting an edge on the competition.

One resource which has been difficult to optimize is Knowledge Management. By Knowledge, we are referring to the intelligence and expertise required to diagnose and repair a technical problem. Knowledge is often considered to be an intangible element of service. IT IS NOT. In many companies, Knowledge is tribal or "referent based" in the sense that it is transmitted from one technician to the other through written (e.g., manuals, notes, etc.) or spoken (i.e., training session) forms. This could be considered an analog approach to KM. As companies become more sophisticated, they often implement digital approaches to KM. Examples range from document management systems at the most basic level of Case Based Reasoning (CBR) on the more advanced. These tools make the end-user more productive and efficient in problem-solving, but they do not necessarily lead to an optimal solution.

To understand why most KM tools are not optimal, we have to examine how they are constructed and operated. Typically, KM tools are populated with information about problems, symptoms, causes, and corrective actions. These databases typically evolve over time. The user or technician will enter the description of the problem or symptom into the software and the application then searches the database to arrive at a possible solution based on previous experience. There are a number of shortcomings with this approach. First and foremost, you may not have had any prior experience with this symptom or problem to affect an optimal solution. Other less obvious shortcomings include:

  • A problem may be the result of multiple symptoms 
  • Multiple problems may produce the same symptom 
  • Multiple symptoms may be the result of multiple problems 
  • The symptom may still appear even after the suspected cause of the problem has been fixed

Basically, CBR offers possible solutions to a problem based on prior history. As CBR tools are implemented, the technician, more or less, takes a trial-and-error approach to resolving a problem. It is true that the tools make the problem-solving process more efficient and thus make the Technician more productive; however, it does not truly optimize the results as to the best solution for treating the symptom and resolving the problem. The shortcoming is that most KM tools take a linear approach to problem-solving. In essence, if the machine is experiencing a symptom, the technician must try solution A to see if it resolves the problem. If that doesn't work the KM suggests another solution. The problem may be resolved for the moment only to reappear a few hours or days later.

Most KM tools help us to find the nugget of gold in our information systems, manuals, or schematics. However, this nugget is buried as one of the several probable answers based on how well we crafted the question. In order to identify this nugget, one needs wisdom, i.e., the intelligence to understand why one solution is better than others. However, "we can be knowledgeable with other men's knowledge, but we cannot be wise with other men's wisdom!" (Michel de Montaigne, French Renaissance author). Let's assume we were going to design the ideal KM tool. We'd probably want it to provide us with the nugget, and not ask the technician to choose wisely from a list of probable solutions!

The reason why most KM solutions adopt a linear approach to problem-solving is that generally speaking, they lack the understanding of the "why" -- which can only come from a deeper understanding of the cause-effect relationships between faults and observations in a complex system. Without it, one can only try to pick the nugget empirically, i.e., evaluate and compare a limited number of options sequentially. But, what if we could evaluate multiple options concurrently based on our understanding of the cause-effect relationships? Would that enable us to pin point the right solution directly in the shortest period of time? That's true optimization. And that is available today!

An Innovative Approach

Model Based Reasoning (MBR) is an innovative approach to KM that incorporates a nonlinear approach to the problem-solving process. MBR makes use of specific data and observations about the current problem to infer or diagnose the root cause. It captures all the ways a device can fail in light of observed symptoms and through a series of complex calculations arrives at the optimal solution. In other words, MBR takes a holistic approach by considering the machine's design, its components, subsystems, and operating conditions, in order to determine the root cause of the problem in real-time.

Without this type of inference engine, Service Organizations often rely on generalizations based on past experience (example: top resolutions for a specific symptom) and rely on the technician to choose the right solution. MBR diagnoses the root cause using a logical process of elimination or reinforcement of possible causes by efficiently gathering and processing information (guided troubleshooting) much like Sherlock Holmes (or, your smartest FSE) would. The fact is the human brain is vastly superior in some respects, and far from being IBMs Watson in others.

Here's a perfect example of how this works. Suppose you observed two symptoms or Error codes on your machine. Let's refer to them as Error Codes "1" & "2". Based on the instrument design we also know there could be four error codes on your machine, showing the following symptoms:

  • Error Code 1 may be caused by A, C, D, E, F, G, H, or I 
  • Error Code 2 may be caused by B, C, D, E, X, Y, or Z 
  • Error Code 3 may be caused by A, B, F, X, Y, or Z 
  • Error Code 4 may be caused by C, D, F, G, H, or I

A smart Field Service Engineer (FSE) will likely consider only those causes that are in common between Error Codes 1 & 2 (namely C, D, and E). They would likely spend time working through C, D, and E sequentially until they arrive at the right solution. In essence, they make several attempts at determining the root cause and resolving the problem. CBR would, of course, order C, D, and E based on past cases, but it would be up to the FSE to determine the root cause. In contrast, MBR would consider the fault codes that are present as well as those that are not and infer the correct solution.

In this example, neither Code 1 or 2 points to a single cause. Each can be triggered by any of 8 faults, of which 3 (C, D, and E) are in common. Further, since there are no observable errors in Codes 3 or 4, any causes associated with these codes can be eliminated. For example, if either C or D were the cause, Error Code 4 would have also been generated. Thus the absence of Error Code 4 proves the absence of C or D, making E the only possible cause that fits the observations. As a result, MBR would arrive at E as the root cause, without having to try C or D. In essence, the diagnostic processes associated with resolving this problem through MBR occurs in parallel (through reasoning) as opposed to sequentially (through trial-and-error). The speed of the root cause analysis is thus much faster than through manual or CBR based diagnostic processes.

Utility of MBR

By utilizing MBR, a service provider can significantly reduce Mean Time to Repair (MTTR) and improve machine availability uptime. It is important to remember that the end-customer is concerned with minimizing machine downtime and ensuring high availability. Downtime can be quite expensive. For example, a semiconductor manufacturer could lose as much as $100,000 for every hour the machine is down. From the customer's perspective, downtime is measured from when the problem is first detected until the machine is placed back into service. The end-user will often attempt to self-diagnose the problem and then seek help from a peer prior to requesting help from the OEM and their FSE. The customer is also likely to run a small test after the FSE arrives on site and corrects the problem. Therefore anything that the OEM can do to reduce downtime either prior to dispatch or while the FSE is on-site will have a positive impact on the customer's operations.

To that end, MBR can be utilized in the following ways to minimize downtime:

  • Guided Troubleshooting (GTS):  MBR provides the end-user with a dynamic set of troubleshooting steps to resolve a problem since it relies on fact-based data as opposed to knowledge history of similar machines in the OEMs installed base.

  • Diagnose before Dispatch (DBD):  An OEM could utilize MBR to analyze problem data reported by the end-user in order to dispatch an FSE with the right skills, parts, knowledge, and tools.

  • Remote Diagnosis (RD): MBR can be used to diagnose problems at a remote location. This is based on the evaluation of data transmitted from instruments that monitor the device and a transfer link to a diagnostic center. RD can be performed through Machine to Machine (M2M), Machine to Person (M2P) or Person to Person (P2P) interface.  

  • Design for Service (DFS):  By considering causes and solutions to potential problems before they have occurred, MBR can anticipate the problem-symptom-cause scenarios that are possible. This information enables an OEM to define a comprehensive service strategy prior to product introduction or to improve the rate at which service issues are detected, isolated and resolved by the service organization.

The quantitative benefits of using MBR in any one of the scenarios identified above are very significant. First of all, by design, MBR has a dramatic impact on reducing the frequency and length of customer downtime. With this improvement comes an increase in customer satisfaction and repeat purchases. Secondly, a service organization can also eliminate a large portion of costs associated with repeat service calls since these types of activities are practically eliminated through the improved accuracy of repairs that are enabled through MBR. Third, MBR plays a huge role in reducing costs associated with service parts logistics. This is possible because MBR eliminates the tendency of FSEs to use service spare parts as a diagnostic tool. As a result, the service organization will experience 1) a reduction in parts usage, 2) a reduction in shipping costs, 3) a reduction in depot repair costs, and 4) reduction in repeat service calls due to installing wrong part.

By integrating MBR into the GTS, DBD, RD, and DFS processes, an OEM can achieve measurable benefits as identified below. These results demonstrate the tremendous impact that MBR has on various OEM stakeholder organizations including service & support, supply chain, sales, and the end-customer. Most importantly these benefits fall directly to both the OEM and the customer's bottom line.

Determining if MBR is right for you

MBR has tremendous benefits when applied to specific types of service environments. Typically this involves situations where their original purchase price of the equipment is 6 figures or more, where the cost of downtime is more than the hourly wages of production personnel, and where the machine operates complex procedures. Here are several questions that you need to consider when evaluating MBR as a KM tool of choice:

√  Does your installed base ever experience service failures that are complex and difficult to diagnose?

√  Do many candidate solutions exist and is troubleshooting required to identify the root cause and corresponding corrective action required?

√  Do you encounter service events where there is no service history and no prior service documentation?

√  Does your company frequently introduce and roll-out new, complex products and

√  Do you need to embrace "Design for serviceability" as a corporate philosophy?

√  Do a majority of your customers require that you provide them with automated or remote diagnostics?

If you answered yes to two or more of these questions, then MBR is likely a good fit for your organization.

Summary & Conclusions

Doing business in today's world of accelerating technological advancement, amid increasing geopolitical and economic uncertainty, regional military conflicts, the effects of climate change on food production, weather induced natural disasters, and a host of other unknowns, present a host of challenges and opportunities to forward thinking companies who are well positioned, and continually repositioning themselves, to meet all contingencies with minimal disruption and maximal efficiency. This is the great test of our time. For those who can keep up by staying ahead of the pack, by minimizing downtime and maximizing productivity, the advantages are significant. To achieve that edge, wise application of knowledge is key. Knowledge combined with wisdom provides actionable intelligence that keeps your systems running.

Significantly bolstering Knowledge Management is Model Based Reasoning which facilitates the aforementioned advantages in an efficient, quantifiable, cost-effective manner. It provides us with wisdom by inferring the right answer from several probable answers. Within specific types of service environments, it provides an innovative approach to KM that is fast, reliable, and flexible, with the ability to grow along with your company, industry, and a rapidly changing world.

At issue, most other KM tools make it faster and easier for you to find the Knowledge within your enterprise. However, these tools do not create knowledge, they only help you record and retrieve it. The key advantage to MBR is that it creates knowledge and accurately records every experience without depending on the user to record or remember the path taken. This is very similar to GPS in your car, you put in the target destination and the system optimizes your route. With MBR you enter the information about Fault and the symptom optimizes your route to the cause. That's true wisdom!

As has been said, "Chance favors the prepared mind." Since first stated well over a century ago by Louis Pasteur, this advice has been repeated many times because genuine words of wisdom have a way of doing that. If as the Chinese suggest, crisis equals danger plus opportunity, then chance and preparedness might be likened to a lighted candlestick and it might be apt to suggest, "Jack be nimble, Jack be quick . . ."

Consultant from
Contact This Sponsor

B2B Solutions Newsletter