disruptions

Hedging against Disruptions with ripple Effects

Daniel Dumke — Mon, 02 Jan 2012 17:10:00 +0000

I have read several articles by Mark Daskin (also reviewed another one here). So with him on the author list of today’s paper I think one can expect a clearcut research question, some kind of mathematical model, a fitting solution method and a definite answer to the underling problem. Well, let’s have a look!

Topic and method

The goal of this research is to analyze disruptions with a regional effect (as opposed to local, single facility disruptions). An example for a regional disruption may be an earthquake, storms or floods. The results should show how to better handle disruptions which are not locally limited and exhibit contagious tendencies.

The authors approach the problem by modeling three separate sub-models, namely the Defender Problem, the Attacker Problem and the User Problem.
The basic assumptions are:

each facility has a fixed capacity,
the capacity is affected by a disruption,
a defense strategy can be defined for selected facilities, it protects the facilities against the disruption

Defender Problem:
“The objective of the defender is to minimize the impact of the disruption by optimally choosing Q facilities to protect. Constraints states that no partial protection of a facility is possible.”

Attacker Problem:
“The objective of this program is to strike the system as hard as possible by identifying the worst-case set of exactly R facilities chosen among those which are not protected (6).2 Finally, no partial interdiction is possible.”

User Problem:
“The user level problem deals with the minimum cost assignment of customer demands to facilities. The total cost is given by the cost for the service provided, plus the penalties paid for the demand which is not met.”

Correlation Matrix:
“The correlation matrix represents the interdependence among the facilities when an attack occurs or, in other words, the extent to which other facilities are affected when a target facility is disrupted.

Figure 1: Map of the 2009 earthquake in L’Aquila, Italy (Liberatore, et al. 2012; USGS)

Solution and Case

The authors develop a solution method, to optimize the given problem.
Data from the 2009 earthquake in L’Aquila, Italy was used to test the model.
Figure 1 shows a map of the area.

The dataset represents the Abruzzo region, a seismically active region of Italy, where L’Aquila is located. The set of the demand nodes, N, includes all the 305 towns and cities in the region and the associated demands, ai, correspond to the number of inhabitants as at beginning of April 2009 expressed in thousands of citizens. The set of initial facilities, F, represents the 35 hospitals of the region.

The model is solved for different parameters in two steps: first calculating the optimal defense program and second calculating the “optimal” attack, also calculating the incurred cost and correlated disruptions.

Figure 2 shows the example for two different solutions:
The buildings represent the available hospitals, stars mark the defended locations and the colored circles represent the disruption, which causes the worst effects.

Figure 2: Two solution Results (a and b). Buildings representing Hospitals, Stars protected Locations, Circles worst possible Disruption Effects (Liberatore, et al. 2012)

Conclusion

The attacker (in this case an earthquake) always looks for the place where it can hit hardest, so the focus is on the worst case scenario: How can one fortify a system so that even in the worst case it still performs at a desired level?

But keeping that in mind also leads to the conclusion that this system might not be for everyone: What if the earthquake does not hit at the worst point? Or not at all? As a rule of thumb, I would say the investments suggested by this model are magnitudes higher than those of an “optimal” (in terms of money) solution. So when do you use such a model?
I think there are two cases: Either if one is dealing with critical infrastructure, like water, food or possibly military installations and (I know its related) if one has to decide how to safe lives: As in the case study itself, where the optimal hospital location has to be found.

Reference:

Liberatore, F., Scaparra, M.P., & Daskin, M.S. (2012). Hedging against disruptions with ripple effects in location analysis Omega, 40 (1), 21-30

Rate This:

Tags:

case study

hospital

protection

worst case

disruptions

Dual Supply Channel Disruption and Supply Chain Design

Daniel Dumke — Mon, 19 Dec 2011 16:35:00 +0000

This is the continuation of the Greece week and today I am going to have a look at a mathematical model to capture the effects of dual disruptions in a news-vendor model.

This time the three authors (Xanthopoulos, Vlachos and Iakovou) come from the Aristotle University of Thessaloniki.

Method

The authors developed a model where a retailer is able to order from two different supply channels. Figure 1 shows the supply chain schematic.
Three different types of the model are modeled separately:

unconstrained model,
fill rate constrained model,
Type I service level constraint,

The decision variables are the order quantities from channel one and two.

Figure 1: Schematics of the Supply Chain Model (Xanthopoulos et al., 2012)

Analysis and managerial insights

The models are analyzed using a numeric analysis with the parameters given in figure 2. Different values for the purchasing cost c_i, disruption probabilities for the supply channels one and two (p_i). The impact of the disruption is denoted by y_i.

Figure 2: Experimentation Data (Xanthopoulos et al., 2012)

144 parameter combinations are analyzed. Figure 2 shows the iso-profit lines for the order quantities Q₁ and Q₂ for different impact levels in the first supply channel (y₁: a) 0.9, b) 0.6, c) 0.1)

Figure 3: Effect of different Disruption Impacts on Supplier 1 (Xanthopoulos et al., 2012)

The authors conclude

It is also observed that as the impact of a disruption on the first channel increases, the optimal solution moves from a solution that mainly utilizes the first supply chain to a solution that mainly utilizes the second one.

Depending on the values of the purchase costs (c_i), the disruption probabilities (p_i), and the effect of a disruption (y_i), one of the suppliers may dominate over the other one, leading to a single supply source.

Nevertheless, for both the uncapacitated and capacitated problems for certain combinations of the values of c_i, p_i, and y_i, a dual-sourcing policy outperforms a single-sourcing one. In such cases, it is optimal to place the larger part of the total order to the dominant channel and its lesser part to the second one, so as to hedge/mitigate the disruption risks.

Moreover, for very high service levels (near 100%), large orders should be placed to both suppliers.

[In] the case in which the first supplier is the dominant supplier as well as the more reliable from the two. In such a case, no matter how high the fill rate will be, orders should be placed only to the first supplier and the second one should not be activated at all.

Conclusion

This article provides some insights into a dual supply channel disruption case. From the title and abstract of the article, I was however expecting more discussion of the results, perhaps the impact of other driving factors beside the impact, probability and ordering cost.

Reference:

Xanthopoulos, A., Vlachos, D., & Iakovou, E. (2012). Optimal newsvendor policies for dual-sourcing supply chains: A disruption risk management framework Computers & Operations Research, 39 (2), 350-357

Rate This:

Tags:

Exception Handling for Robust Supply Chain Design

Daniel Dumke — Wed, 07 Dec 2011 18:33:00 +0000

This article considers the design of robust supply chains from the viewpoint of exception handling. Disruptions of the supply chain happen all the time. Smaller disruptions like quality issues are part of the daily business; but a look at the recent ten years shows that large disruptions, happen more often as well (think of terrorist attacks or earthquakes). “Thus, exception management is an important issue in global supply chain networks.”

Basics and definition

If one accepts the fact, that exception cannot be eliminated completely there are two ways to treat them: preventive and interceptive. Figure 1 shows different exception management strategies.

Figure 1: Exception Management Strategies (Gaonkar and Viswanadham, 2003; click to zoom)

The authors define a supply chain exception as the opposite to the classical seven-Rs of logistics:

We can use this description to define a supply chain failure or exception occurring whenever the supply chain deviates from any one of the […] required specifications – either in terms of delivering the wrong product, in the wrong quantity, in the wrong condition, at the wrong place, at the wrong time, at the wrong cost and to the wrong customer.

Exception management

In this case exception management is addressed “at the strategic level through the preventive selection of supply chain partners that mitigate risk in the network.”

In the first step the consequences of a given exception (here: supplier non-performance) are analyzed. The authors use a Cause-Consequence-Diagram for this end (figure 2).

Figure 2: Cause and Consequence Diagram for one Example (Supplier Non-Performance) (Gaonkar and Viswanadham, 2003; click to zoom)

Then, “given the probability of occurrence of the initiating event, which is supplier non-performance, and the probabilities for the various intermedialy events, we can calculate the probability of occurrences for each of the end states or outcomes. Furthermore, each of these end states may result in different levels of supply shortfalls and financial cost. Hence, given the probability of each end state and the supply shortfall or financial cost for each end state, we can calculate the expected shortfall or financial risk for the non-performance of a given supplier.”

If repeated for each supplier, the supplier with the least expected impact can be identified.

Model

To implement their approach the authors develop a small linear mixed-integer model.

The model was formulated in Microsoft Excel and solved using the Solver add-in. The model was solved for a problem with a single manufacturer (located in the US), dealing with 5 suppliers. The probabilities of supplier disruption for all the suppliers (individually and in various combination) were considered as given [the first part is shown in figure 3]. The relation cost was taken as $5000 and the quantity required by the manufacturer was 520 units.

Figure 3: Scenario / Probability Table (Gaonkar and Viswanadham, 2003)

Conclusion

The presented approach makes it easy to include uncertainty into the supplier selection process. Considering the moderate efforts necessary for implementing the supplier assessment and implementing the model, I would like to read more about the effectiveness in a real business situation.

Reference:

Gaonkar, R., & Viswanadham, N. (2003). Robust Supply Chain Design: a Strategic Approach for Exception Handling Proceeings of the 2003 IEEE International Conference on Robotics & Automation, 1762-1767

Rate This:

Tags:

Predicting and Managing Supply Chain Risks

Daniel Dumke — Mon, 21 Nov 2011 15:40:00 +0000

I often recommend Supply Chain Risk: A Handbook of Assessment, Management, and Performance as a good overview of supply chain risk management literature. In this article I summarize chapter four of the book on the problem and processes to predict supply chain risks.

Structure and basics

In first section the author starts by defining the basics of risk and uncertainty. The next section briefly introduces different categories for supply chain risks:

Internal to the firm: Process, Control

External to the firm but Internal to the Supply network: Demand, Supply

External to the network: Environmental

These risk can then impact the different dimensions of a supply chain:

Physical movement of goods
Flow of information
Flow of money
Security of the firm’s internal information systems
Relationship between supply chain partners
Corporate social responsibility and the effect on a firm’s reputation.

The next section provides an overview on supply chain risk management, its steps and mitigation strategies.

Predicting Supply Chain Risks

One of the main requirements for an effective proactive risk management process is to obtain good estimates of the probability of the occurrence of any particular disruption and accurately measure the potential impact.

So the estimation of probabilities and impacts, depending on the characteristics of the supply chain and the environment it might be harder to identify all risks and assess them.

The author describes two tools for the estimation of risks: Data Mining and Failure Mode Effect Analysis (FMEA).

Data Mining

The role of data mining is to analyse historical data, to improve prediction capability. Some of the common analytic approaches used by data miners are:

Estimation of the parameters of past performance: Means, Standard deviations, Correlations, and Associations for hypothesis testing

Classification: Segmentation, or Clustering of data units to facilitate the modelling process

Construction of a functional relationship: or model between responses and explanatory variables.

The following tools should be used by a data miner:

Estimation: […] These tools will not lead to patterns but are more useful in analysing the data to identify the most relevant sets of data to concentrate further analysis.

Clustering/segmentation: This approach is used to logically group observations on the basis of similarity in their characteristics, reducing the level of heterogeneity in the data.

Classification/discrimination: The process of assigning observations to a predetermined number of classes.

Prediction: Formal mathematical models are built for the purpose of predicting the occurrence of the phenomenon.

Failure Mode Effect Analysis (FMEA)

The goal of the FMEA is to

Recognise and evaluate the potential failures of a product or process and the effects of that failure.

Identify actions, which could eliminate or reduce the chance of the potential failure occurring.

Document the entire process.

The author highlights ten steps of the process:

Define scope, functional requirements, design parameters and process steps.

Identify potential failure modes: Failure modes indicate the loss of at least one functional requirement. It is the manner in which a failure occurs. This step in the process takes into account a foresight view (based on past experience and any new information) of what could cause a failure to the system or process.

Potential failure effect: This step investigates the effect the failure will have on other entities or processes.

Severity: “How bad” or “serious” the effect of the failure mode is. Usually severity is rated on a discrete scale from 1 (no effect) to 10 (hazardous effect). Severity ratings of 9 and 10 indicate a potential effect of high importance and this could typically be a safety or government regulation issue. Critical effects need deeper study for all causes to the lowest level, using a method of Fault Tree Analysis.

Potential causes: These are the causes of the failure. In this step, all causes that can be attributed to the failure occurring are investigated.

Occurrence:This is the likelihood of the event happening (i.e. failure in the system) on the basis that “the cause occurs”. FMEA assumes that if the cause occurs, failure will occur too. The probability of occurrence is ranked from 1 to 10, where 1 signifies a remote probability of occurrence and 10 a very high probability of occurrence.

Current controls: The objective of the controls is to identify and detect the deficiencies and vulnerabilities as early as possible. This step looks at the current processes in place to mitigate the failures (if already known).

Detection: A subjective rating is assessed corresponding to the likelihood that the detection method will detect the first-level failure of a potential failure mode. This is ranked from 1 to 10, where 1 signifies that it is unlikely to detect and 10 signifies a very high detection potential.

Risk Priority Number (RPN): These are used to prioritise the potential failures and are calculated as “Severity Occurrence Detection ranking”.

Actions Recommended: The team should then select and manage subsequent actions needed to locate and control the situation.

Predictive – Proactive Methodology

Based on these methods the author suggests a predictive-proactive methodology (figure 1) to manage risks.
The two main steps are called predictive mode and proactive mode. In the predictive mode, the focus is on gathering of information on the potential risk sources. The proactive mode focusses on the identification of the risks within the supply chain and assessing the risk level and setting the strategic objectives of the company.

Figure 1: Schematic Diagram of the Predictive-Proactive Approach Supply Chain Risk Management Approach (Dani, 2006)

Conclusion

I read the chapter with a focus on the description of the prediction tools. The model on the other hand is a nice add-on which could be used as food for thought. But I find it a little bit too simplistic and a little bit confusing. It describes a basically linear process (from a risk to the disruption), but why then did the author choose not to represent this in a linear way, but in a diagram with two main parts/frames instead of a clear process and why are there two different types of frames (dashed and solid)?

Reference:

Dani, S. (2006). Predicting and Managing Supply Chain Risks Supply Chain Risk (Ed. Clare Brindley), 53-66 DOI: 10.1007/978-0-387-79934-6_4

Rate This:

Tags:

forecast

risk

disruptions

Ways to guard against Disruptions

Daniel Dumke — Mon, 17 Oct 2011 17:49:00 +0000

Today I review an article called “18 Ways to Guard Against Disruption”. It was published in the Supply Chain Management Review in 2005 by Elkins et al.

Goal and method

The goal of this study was to assess the current state of supply chain risk management capabilities across multiple industries. The authors describe the results of several interviews with companies from multiple industries, which have been conducted by the Supply Chain Resource Consortium (SCRC, an university industry partnership for documenting supply chain management knowledge).

Results

Based on this study 18 best practices have been identified, which have been categorized according to their impact location (internal or external) and their time orientation (current business vs. future business).

External orientation / future business (“Strategic sourcing & advanced procurement”)

“Screen and regularly monitor current and potential suppliers for possible supply chain risks.
Require critical suppliers to produce a detailed disruption-awareness plan and/or business-continuity plan.
Include the expected costs of disruptions and operational problem resolution in the sourcing total-cost equation.
Require suppliers to be prepared to provide timely information and visibility of material flows that can be electronically shared with your organization.”

External orientation / current business (“Supply-base management”)

Conduct frequent teleconferences with critical suppliers to identify issues that may disrupt daily operations and discuss tactics to minimize these issues.
Seek security enhancements that comply with the Customs-Trade Partnership Against Terrorism (C-TPAT), Container Security Initiative (CSI), and similar initiatives.
Test and implement technologies to track containers.
Conduct a detailed incident report and analysis following a major disruption.
Create exception-detection/early-warning systems to discover critical logistics events that exceed normal planning parameters.
Gather supply chain intelligence and monitor critical supply-base locations.

Internal orientation / current business (“Real time operations management”)

Improve visibility of inventory buffers in domestic distribution channels at the part level.
Classify buffered material by its level of criticality.
Train key employees and groups to improve real-time decision-making capabilities.
Develop decision-support tools that enable the company to reconfigure the supply chain in real time.

Internal orientation / future business (“Strategic supply chain design”)

Develop predictive analysis systems that incorporate intelligent search agents and dynamic risk indices.
Construct damage-control plans for likely disruption scenarios.
Understand the cost trade-offs for different risk-mitigation strategies.
Enhance systemwide visibility and supply chain intelligence by using improved near-real-time databases.”

The authors note that none of the companies had implemented all of the identified measures at the same time and they suggest:

Companies may want to use the best-practices list as a thought-starter to help them prioritize which supply chain risk management elements to adopt. For example, companies could develop an internal survey, based on the best-practice list, which would assess their supply chain risk management capabilities.

Conclusion

This article closes a gap at the SCRM Blog of some early exploratory research on how supply chain risk management is done in practice. Even though the description of the methodology lacks several important facts, e.g. how many companies have been interviewed and how were the interviews conducted, the results still show an interesting direction.
The next obvious questions, which to my knowledge have not been pursued in an systematic manner, would be:

Which of the strategies is best used in what setting?
How should a company prioritize their efforts, if they wanted to invest in supply chain risk management?

Reference:

Elkins, D, Handfield, R.B., Blackhurst, J., & Craighead, C.W. (2005). 18 Ways to Guard Against Disruption Supply Chain Management Review

Rate This:

Tags:

disruptions

mitigation

risk management

Managing Disruption Risks using Real Options (SCRM Thesis)

Daniel Dumke — Mon, 26 Sep 2011 13:17:00 +0000

This is the seventh contribution to my series on doctoral and master dissertations on Supply Chain Risk Management. This again is a master thesis from the MIT. An immense effort and dedication is spent on these works only to find the results hidden in the libraries. So the goal is raise interest in the research of my peers.

Author / Topic

This thesis was written by Sophie Pochard already in 2003 as her master thesis at the Massachusetts Institute of Technology, Cambridge, USA. It can be downloaded here directly at the MITs web site. It has been supervised by Richard de Neufville, Professor of Engineering Systems and Civil and Environmental Engineering. The title is: Managing Risks of Supply-Chain Disruptions: Dual Sourcing as a Real Option.

Summary

In her thesis Pochard focussed on multi-sourcing as sole risk mitigation strategy. But first the author elaborates on how resilience in a company can be achieved and more interestingly some limitations to resilience with a focus on dual sourcing, which are:

Complexitiy, increasing resilience often also increases the complexity of the supply chain (eg. dual sourcing).
Risk trade offs, having only a single supplier can also reduce risks (eg. protection of intellectual properties).
Cost issues, resilience often increases cost.

To analyze multiple sourcing strategies Pochard then introduces real options, where the definition is similar to that of a financial option contract:

Call options give to their holder the right to undertake an investment, at a cost that is fixed in advance (the exercise price), at or before a given date (the maturity). For example keeping unexploited leases can be assimilated to a call option. Oil companies do not sell or exploit these leases as they want to keep the right to develop them later on; they may, for example, decide to use this right if ever new drilling and production technologies allow to increase recoverable reserve. This is a call option since the exploitation makes it possible to get the income of the underlying. Its exercise price is the investment cost to initiate the production, and the maturity is simply the date until which firms have the authorization to exploit.

So as the next step dual sourcing is modeled as a real option. And the author presents her results in several diagrams where I show three here in the blog (figures 1 and 2).

Figure 1: Strategy Selection depending on Disruption Frequency (Pochard, 2003)

Figure 2: Sensitivity Analysis based on the Price of the local Supplier (Pochard, 2003)

Both independent variables (disruption frequency and cost of local supplier in case of disruption) seem to have a major impact towards the selected strategy. Pochard concludes that the real options can be a valid method to analyze the benefits of different strategies.

Conclusion

I think that her thesis can be used as a basis for a variety of topics. She writes about disruption risks, resilience and risk mitigation strategies, of course the definition of real options and how to develop and deploy a corresponding model.
I also think that this variety is also the major weakness of this work. Compared to my last MIT master thesis (Impact of demographics on supply chain risk management practices) Pochard takes a very long time to get to the real topic and browses a long time through other connected fields.
Nonetheless the results are very interesting and seem to be comparable to other (later) papers.

Reference:

Pochard, S. (2003). Managing Risks of Supply-Chain Disruptions: Dual Sourcing as a Real Option Massachusetts Institute of Technology, Master Thesis

Rate This:

Tags:

Managing Supply Chain Disruptions in the Chemical Industry

Daniel Dumke — Wed, 14 Sep 2011 17:19:00 +0000

This is already my second article (click here for the first) about managing supply chain risks in the chemical industry. This time by Paul R. Kleindorfer and Germaine H. Saad from Wharton and the Widener University. But this industry is quite interesting since it has to withstand a multitude of risks, so let’s get right to business:

Goal

As the title indicates: the authors focus on risks from disruptions to normal activities and not the omnipresent risk of coordinating supply and demand.
The authors aim to achieve three goals:

Development of a conceptual framework for risk assessment and mitigation
Analyze accident history from the six years between 1995 and 2000
Derive implications for the design of management systems to cope with supply chain disruptions

Figure 1: Risk Management Process in the Extended Supply Chain (click to enlarge; Kleindorfer and Saad, 2005)

Literature analysis

The authors use the SAM method, where the goal is first to analyze the risk S ources, then A ssess the risks and find the right M itigation strategies.
The authors first investigate the SAM process using literature analysis and conclude with the process depicted in figure 1 and a four step process to manage disruption risks:

“Obtain senior management understanding and approval, and set up organizational responsibilities for managing the Disruption Risk Management process.”
“Identify key processes that are likely to be affected by disruptions and characterize the facilities, assets, and human populations that may be affected.”
“Traditional risk management is then undertaken for each key process to identify vulnerabilities, triggers for these vulnerabilities, likelihood of occurrence, and mitigation and risk transfer activities.”
“Reporting, periodic auditing, management, and legal reviews of implementation plans and on-going results (e.g., of near-miss management and other disruption risks) complete the business process for disruption risk management.”

Case study chemical industry

The further analysis of disruption risk management is done using a case study approach. The authors used a database with access to accident activities in the chemical industry during the time between 1995 and 2000. Sources were natural hazards, equipment or human failure and sabotage.

The framework depicted in figure 2 was used as a basis for the analysis.

Figure 2: Research Framework for the Analysis of the Chemical Industry (Kleindorfer and Saad, 2005)

Results

Kleindorfer and Saad conclude that, “two key dimensions emerge as fundamental in guiding management practice of disruption risk in supply chains:”

The first dimension consists of strategies and actions aiming at reducing the frequency and severity of risks faced, at both the firm level and across the supply chain. The second element focuses on increasing the capacity of supply chain participants (whether a separate firm or a subsidiary facility) to sustain/absorb more risk, without serious negative impacts, or major operational disruptions.

Furthermore in the chemical industry special care has to be taken to include the stakeholders, like regulators or the employees in mitigation activities.

Lastly, the authors state two conditions which have to be met to ensure effective implementation of risk management measures:

Condition C1. The approaches used to mitigate disruption risks must “fit” the characteristics and needs of the underlying decision environment. Different supply chain environments will give rise to different approaches to assessment and mitigation. Even within a given sector such as mass retailers with global supply chains, the approaches taken by different companies will be different, given their differences in corporate culture and their methods of managing their supply chains. In the business of disruption risk management, one size will definitely not fit all.

Condition C2. Trust, information and continuing profitability are the basic glue that makes supply chain partnerships a reality. Continuous coordination, cooperation, and collaboration among supply chain partners are needed for risk avoidance, reduction, and mitigation such that the value and benefits generated are maximized and shared fairly. Different contractual and incentive schemes are now emerging as practical means toward this end. Linking risk assessment and risk management to identify vulnerabilities and opportunities for cost-effective mitigation is the first step in providing a rational basis for individual and group action among supply chain partners. Random investments or shots in the dark that do not properly account for the interdependencies across the supply chain will ultimately waste resources and destroy trust.

Conclusion

The authors already admit that the data used for their case study is based on a facility level. These facilities are of course part of the supply chain, but one can pose the question if the results are really relevant for supply chain risk management.

Moreover, one has to differentiate between supply chain security and supply chain disruptions. At least from my practical experience especially in the chemical industry facility security has played a major role since many decades and is deeply rooted in the companies themselves. Supply chain risk management on the other hand is a new approach and very much separated from it. So for a paper on exactly this topic I am missing a discussion of the question how these aspects should be integrated.

Reference:

Kleindorfer, P.R., & Saad, G.H. (2005). Managing Disruption Risks in Supply Chains Production and Operations Management, 14 (1), 53-68

Rate This:

Tags:

Let me help you with... Time-Based Risk Management

Daniel Dumke — Wed, 31 Aug 2011 13:56:00 +0000

Today I present you only one chapter of a great book by Wu and Blackhurst: “Managing Supply Chain Risk and Vulnerability” (which can be bought on amazon.com here).

I especially like the third chapter by Professors Sodhi and Tang on a new approach to managing disruptions: Time-based Risk Management.
The whole book features a more practitioner oriented approach so don’t expect too much focus on the methodological parts.

Reason for risk management

Supply chain disruptions should have arrived at the board rooms due to

“short-term effects (negative publicity, low consumer confidence, market share loss, etc.) and
long-term effects (stock prices and equity risk).”

So there should be enough drive to pay for even some of the more expensive risk management initiatives. Nonetheless a really robust supply chain strategy has to mitigate supply chain risks and increase profits at the same time.

Time-based risk management

The time-based risk management approach aims to travel on this thin line and deliver a strategy to mitigate disruption risk without compromising profits.
It consists of three time frames which should be in focus of the risk manager:

time to detect a disruption (D1),
time to design a solution (D2),
time to deploy (D3),
time to response is set to the sum of D1 to D3 (R1), and
time to recover (R2)

The authors argue, that prior work has pretty much focussed on the generation and selection of recovery plans, but only after the event has occurred.
The time-based risk management now tries to reduce the time needed for the other elements, since “just as 80% of the total cost of a product is determined during the product design phase, the activities that take place for designing response can have significant effect on the overall impact of a disruption.”

Argumentation

Furthermore, based on three case studies (e.g. the Nokia / Ericsson Case) the authors argue that a longer response time can also lead to hugely increased recovery times. “This is mainly because the magnitude of the problem triggered by the event escalated exponential over time.”

This argument is illustrated in figure 1 using a exponential epidemic model.

Figure 1: Effects of Response Time Reduction (Sodhi and Tang, 2009)

Activities to reduce response time

The authors name five time-based disruption management strategies to reduce the response time (R1) and therefore also the recovery time (R2).

Work with suppliers and customers to map risks
Define roles and responsibilities
Develop monitoring/advance warning systems for detection
Design recovery plans
Develop scenario plans and conduct stress tests

Risk and reward

As argued in the beginning robust strategies should not only support the company during times of disruption, but also increase competitiveness.
Using the case of Zara (fashion industry) the authors outline the advantages of time-based risk management. By managing their design, manufacture and deliver operations very closely and centralized with a already very low lead time Zara can detect disruptions almost in real time (low D1). New clothes and therefore a market adaptation can be designed very quickly, so D2 is quite low as well. Since suppliers and distribution are usually in close proximity to the factories deployment of new strategies is fast as well (low D3).

However these low lead times have not (primarily) been established for risk management purposes, but they are part of Zara’s business strategy and help them achieving continuous high growth rates as well.

Conclusion

Sodhi and Tang use very illustrative cases to make their points for a time-based risk management approach. This approach represents a corporate strategy which of course has to include a very broad definition of supply chain management. So not only the logistics and manufacturing parts are included, but also product design, finance, …

Reference:

Sodhi, M.S., & Tang, C.S. (2009). Managing Supply Chain Disruptions via Time-Based Risk Management Managing Supply Chain Risk and Vulnerability: Tools and Methods for Supply Chain Decision Makers., 29-40

Rate This:

Tags:

Pricing in Times of Disruption

Daniel Dumke — Wed, 10 Aug 2011 16:13:00 +0000

Many articles, including my own research show, that companies tend to focus largely on risk mitigation measures concerning the supply side. Only little is done to include demand side risks or demand side measures into the mitigation of supply chain risks. The study “Pricing During Disruptions: A Cause of the Reverse Bullwhip Effect” focusses on optimal pricing measures during a disruption. And so it helps to close the gap a little bit.

You can download a preprint of today’s paper at SSRN.

Reverse Bullwhip Effect

During a disruptions demand can change quickly. Due to a real or felt shortage customers are likely to order more than they actually need. This effect can be described by the Reverse Bullwhip Effect:

Whenever there is a perceived shortage of supply, it amplifies as it propagates down the supply chain. In fact, forward and reverse bullwhip effects often act as a system. If you start with a sudden upturn in demand, it gets amplified as it goes upstream, which creates a perceived shortage that amplifies as it propagates downstream. This creates a panic and downstream consumers overstate their demand, which amplifies again as it goes upstream. In turn, a greater scarcity is felt, and so on. Each of these effects feeds the other.

Method

The authors establish a mathematical model of a two tier supply chain, containing a single manufacturer (M) and a single (aggregated) demand (C). The process is established as follows:

Before the interaction, the manufacturer is hit by a disruption with a decreasing effect on capacity. The recovery takes place slowly over the course of several periods. The following steps are executed in each period:

M realizes its current capacity

M sets a price

C orders a specific number of products according to its predetermined behavior

The manufacturer has three price setting strategies at its disposal:

Naive: The price is set only based on the capacity and the long term demand curve of the customer, any short term behavior of the customer is ignored
One Period Correction: The deviation between the long term demand and the actual order of the customer is taken into account for the pricing decision.
Regression pricing: A regression analysis between set price and customer orders is included into the decision making process of the manufacturer.

The customer’s order behavior also includes historical prices. Two demand curves are specified, a long term demand curve, where the demand is linear in price, and a short run demand curve where the price change is included as well (figure 1).

Figure 1: Short Term Demand Behavior (Rong et al. 2009)

Results

The results show that the one period correction strategy results in a more volatile customer ordering process and lower revenues than both the naive and regression pricing strategies. Also in these terms the regression pricing strategy performs

worse than the naive strategy. The Reverse Bullwhip Effect is shown to occur in the disruption setting and almost always leads to reduced revenues.

Conclusion

From my point of view the study emphasizes four things:

All strategies involve active price changes by the company. Even though the naive strategy sounds very “lazy” it still involves using the available knowledge to optimize the profits of the manufacturer, including adapting prices.

The limited knowledge of the manufacturer is the major obstacle in the reduction of the Reverse Bullwhip Effect and improved revenues.

As for the Bullwhip Effect itself, the effects might be even worse with longer supply chains

An open question for me would also be how a more realistic (i.e. slower) price setting algorithm would affect the strategies.

Reference:

Rong, Y., Shen, Z.-J. M., & Snyder, L V. (2011). Pricing During Disruptions: A Cause of the Reverse Bullwhip Effect SSRN

Rate This:

Tags:

Recurrent versus Disruption Risk

Daniel Dumke — Mon, 18 Jul 2011 13:25:00 +0000

The distinction between disruptions and recurrent / continuous risks is commonly used by researchers and practitioners in supply chain management. But how should the differences be reflected in the supply chain planning process? Is it necessary to differentiate between the risk types here as well?

In 2007 Sunil Chopra et al. analyzed this question in depth.

Disruption vs. continuous risks

For simplicity sake the authors define a disruption as a supplier which does not deliver any product at all. Figure 1 shows an example delivery listing for 20 periods. In periods 4, 7 and 13 nothing has been delivered and so a disruption has happened.

As you can see the overall mean (including the disrupted periods) is 86 with a standard deviation of 39. But the third column highlights an important point: If you leave out the disrupted periods, the mean rises and the standard deviation drops to a third. As a result it should make a difference if a decision maker used one or the other mean / standard deviation to make his decision.

Figure 1: Exemplary Table containing Recurrent and Disruption Risks (Chopra et al., 2007)

Model

To test the effects of a differentiated versus a bundled view the authors develop a model. The focal company has two suppliers, one reliable and one unreliable (at risk of a disruption). The authors then analyze two cases: first, only recurrent risk and second, recurrent and disruption risks. Each with either a decision maker who views the risks bundled together or separately.
Figure 2 has a good graphic to show the implications of the different approaches with disruptions present. In a bundled view the excess order volume (y-axis, to prevent stock-outs) is increasing when the probability of the disruption increases. On the other hand, when the risks are decoupled the decision maker only focusses on the recurrent risk and the excess order even decreases with the probability of a disruption. The disruption risk is then mitigated by contingently using the more reliable supplier.

Figure 2: Comparison of Excess Orders for Decoupled and Non-Decoupled Risks (Chopra et al., 2007)

Implications

The authors conclude that:

Bundling of disruption and recurrent supply uncertainty results in an over (under) utilization of the low cost (reliable) supplier. The extent of over (under) utilization increases as the probability of disruption grows.

Furthermore:

Growth in supply risk from increased disruption probability is best mitigated by increased use of the reliable (though more expensive) supplier and decreased use of the cheaper but less reliable supplier. Growth in supply risk from increased recurrent uncertainty, however, is better served by increased use of the cheaper, though less reliable, supplier.

Conclusion

I liked the article a lot, since it focusses on one simple question: Is it bad to combine different types of risk. The conclusion: There will always be different risk sources or types which, as this article shows, have to be mitigated differently as well.

Reference:

Chopra, S., Reinhardt, G., & Mohan, U. (2007). The importance of decoupling recurrent and disruption risks in a supply chain Naval Research Logistics, 54 (5), 544-555 DOI: 10.1002/nav.20228

Rate This:

Tags:

Impact of Disasters on different Sectors

Daniel Dumke — Wed, 13 Jul 2011 11:04:00 +0000

I already wrote about different effect supply chain disruptions can have on a focal company and its stakeholders. Now I found another interesting article dealing with the impact of different disasters on different industries within the supply chain.

Method

The authors (Altay and Ramirez) use a exploratory empirical study to analyze the effect of over 3’500 historic natural disasters within over 150’000 firm-years.
From the literature analysis three hypothesis were generated:

H1: A firm’s financial leverage increases in response to a disaster.
H2: A firm’s TAT (Total Asset Turnover) will decrease in response to a disaster.
H3: A firm’s OCF (Operating Cash Flow) will increase in response to a disaster.

Two disaster databases were used to gather the relevant data about the disruptions. Three proxies for the overall effect of the disaster are used: ratio of damage over GDP, affected people by the disaster (per capita) and a composite measure based on the disaster count, affected population, the death toll and the damage.
These were then correlated with the financial leverage, the TAT and the OCF.

Results

Regarding their hypothesis the authors follow for all industries

“Results using our monetary proxy suggest a positive and time-persistent correlation between financial leverage and disaster damage for all sectors with the exception of extractive industries.”
“In general, firms become less efficient in managing their assets (lower sales as a percentage of assets) after a disaster.”
“When we turn to our composite measure, we see support for the negative impact of disasters on firm cash flows.”

The results are nonetheless not consistent over all proxies (damage, people affected and composite), so the authors follow rightly, that the correct proxy has to selected very carefully.

The authors go further in their data analysis and find that…

…damage by windstorms and floods seem to be dramatically different from that of an earthquake, providing evidence against the all-hazards approach.

furthermore:

We also show that the impact of floods on TAT of a firm is dependent on the firm’s position in the supply chain. We found that while upstream partners enjoy a positive impact, downstream partners have to plan for the opposite.

That last finding is quite interesting, since it shows that floods (and windstorms) seem to be disruptions which can be planned for and stock is accrued in advance and sold successfully.

Conclusion

The authors also take some effort to support the robustness of their results. The overall results remain unchanged after removing:

larger countries (with potentially more disruptions by which not all companies are affected equally),
G8 countries, and
countries with higher insurance consumption.

Overall I found this article kind of hard to read. Probably because the reading flow is disrupted by several multiple page tables which could have been moved to the appendix.
But still the article takes a new approach to evaluating the effect of supply chain risks on companies and industries and brings some interesting findings.

Reference:

Altay, N., & Ramirez, A. (2010). Impact of Disasters on Firms in Different Sectors: Implications for Supply Chains Journal of Supply Chain Management, 46 (4), 59-80

Rate This:

Tags:

Planning for the Catastrophe

Daniel Dumke — Mon, 11 Jul 2011 09:14:00 +0000

Practitioners often complain about the huge gap between practice and research related to the estimation of risks. In theory all is easy: A disruptive event just gets a probability and outcome assigned. But in practice these figures most often have to be estimated.
Todays article by Knemeyer et al. (2009) covers exactly this dilemma and tries to answer the question of how to plan for a catastrophe.

Catastrophes

A catastrophe can be defined as an event which happens with a low probability but comes with a large impact on the business.

Figure 1: Defining Catastrophes (Knemeyer et al., 2009)

Estimating risks

Risk analysis is a major part of risk management. Figure 2 shows an example of a process where the data from a risk assessment is combined with the understanding of how risks are perceived by managers (e.g. read my articles on Managerial Risk Perception Part 1 and 2). Depending on the context managers seem to either over- or underestimate the probability of low-likelihood events, and since accuracy is very important (if you miss judge the probability of an 1% event by +1% you already doubled your probability and with it the expected loss)

Figure 2: Process for Risk Analysis (Knemeyer et al., 2009)

In the next step the strategies have to be devised and selected. Finally, the strategies should be evaluated on a regular basis to foster organizational learning.

Planning process

Figure 3 shows the suggested planning process.

Figure 3: Process view: Planning for Catastrophes (Knemeyer et al., 2009)

The first step in the planning process is to identify key supply chain locations. A location is considered key if interruption of its operations results in a major disruption in the flow of goods in the supply chain.

After finishing the first step, the manager should have a list of key locations, together with the major catastrophic events that should be considered.

As a next step the probabilities have to be estimated. The authors suggest catastrophe simulation as a tool to estimate probabilities. Figure 4 shows other possibilities. In addition to that, the potential losses have to be evaluated as well.

Figure 4: Methods used for Probability Estimation (click to enlarge; Knemeyer et al., 2009)

For the third step potential countermeasures are designed and selected in the next step for each key location. Figure 5 shows a sample of countermeasures and their impact on probability and incurred loss.

Figure 5: Sample Strategies to mitigate Catastrophes (click to enlarge; Knemeyer et al., 2009)

To select the strategies the authors suggest simple expected loss comparisons.

Conclusion

The suggested process should be suited for implementation within a company. Of course there are several simplifications which have to be analyzed prior to implementation. For instance, the authors suggest to use scenarios (each with probability and impact). I believe that when starting a process implementation like this it might be sufficient to use plain probabilities, but later on probability distributions make show a more detailed and accurate picture. But, I am wondering what you think? Could a process like this be used to improve the risk profile of a company?

Reference:

Knemeyer, A., Zinn, W., & Eroglu, C. (2009). Proactive planning for catastrophic events in supply chains Journal of Operations Management, 27 (2), 141-153 DOI: 10.1016/j.jom.2008.06.002

Rate This:

Tags:

Single or Dual Sourcing

Daniel Dumke — Mon, 04 Jul 2011 14:10:00 +0000

Today there is only one simple question: Single or dual sourcing, what strategy should be used when faced with disruption risks.

Definitions

There are four different approaches to sourcing sole, single, dual and multiple sourcing. Sole sourcing refers to a scenario where there is only one supplier in the supplier base, whereas single sourcing refers to the conscious decision to select one, single supplier out of a choice of suppliers. The dual sourcing model indicates that two suppliers are used (not necessarily at a comparable share). As the name says with multiple sourcing, multiple suppliers are selected, depending on the price or another criteria. The paper presented here focusses on single and dual sourcing only.

Model

The authors (Yu, Zeng and Zhao) make use of a mathematical model which contains a two echelon supply chain with two suppliers. One of the suppliers (main supplier) is located offshore (long lead times) and prone to disruptions, but on the other hand less expensive. The second supplier is located locally, but more expensive. Therefore there are three different sourcing alternatives: single sourcing of the main or the secondary supplier or alternatively dual sourcing with both suppliers.

Demand in the model is correlated with the price and stochastic.

Results

For each of the sourcing strategies the authors prepare and compare the profit equations. As a result the authors conclude that there are two critical disruption probabilities at the main supplier: p₁^c and p₂^c (p₁^c2^c) which build a frame for three different cases:

Case 1: the disruption probability of the main supplier is less than p₁^c then it is optimal to only source from the offshore main supplier.
Case 2: p₁^cdisruption 2^c, then it is optimal to source from both sources
Case 3: the disruption probability is larger than p₂^c, then it is optimal only to source from the reliable, local supplier.

Figure 1 shows the three cases.

Figure 1: Profit vs. Disruption Probabilities (Yu et al., 2009)

Conclusion

Of course the formulas in the model could be used to calculate the above mentioned probabilities and then compare them to the current disruption probability to decide on the optimal strategy. But this would probably lead to wrong results, since the model contains many assumptions that are probably not met in your case study. But the intension of the model is clear: To highlight that depending on the probability of a disruption it, not only can be optimal to have dual sourcing but also to completely switch to a more reliable supplier to optimize profits.

Reference:

Yu, H., Zeng, A., & Zhao, L. (2009). Single or dual sourcing: decision-making in the presence of supply chain disruption risks Omega, 37 (4), 788-800 DOI: 10.1016/j.omega.2008.05.006

Rate This:

Tags:

Avoiding Supply Chain Breakdown

Daniel Dumke — Mon, 20 Jun 2011 14:00:00 +0000

This article presents a comprehensive practice oriented framework for managing supply chain disruptions by Sunil Chopra and ManMohan S. Sodhi. The article has been published in the MIT Sloan Management Review in 2004. The framework covers everything from risk analysis to the selection of the risk mitigation strategy.

Risk Categories

Chopra and Sodhi find nine categories for risk in the supply chain context. In figure 1 they contrast the risks with the corresponding risk drivers.

Figure 1: Risk Categories (Chopra and Sodhi, 2004)

Risk mitigation

The above mentioned risks have to be analyzed and if necessary mitigated. Risk mitigation as a process of managerial decision making can be quite cumbersome, since there is no catch-all strategy and some strategies can even increase other risks as a side effect. Figure 2 highlights the effect of selected mitigation strategies on the risk categories.

Figure 2: Effect of Risk Mitigation Strategies (Chopra and Sodhi, 2004)

Cost of risk mitigation

Very often risk mitigation not only comprises a laborious decision process, but also the resulting strategy implementation can mean a huge financial expenditure. Some strategies on the other hand are not only capable of reducing risks but can also increase profitability. Figure 3 highlights the dilemma. The goal should be to jump from the current position “X” to a higher risk-return curve (eg. “A”), instead of moving within the existing decision space.

Figure 3: Trade-off between Risk Reduction and Cost/Reward (Chopra and Sodhi, 2004)

What-if analysis

The authors suggest to use a what-if scenario analysis to assess a companies / supply chains propensity towards certain risks. In the matrix displayed in figure 4 the risk categories are further separated into supplier-related, internal and customer-related parts.

Figure 4: Exploring Scenarios using What-If Analysis (click to enlarge; Chopra and Sodhi, 2004)

Risk and reward

To balance risk versus reward the authors recall three essential relationships:

The cost of risk reduction increase with the level of risk (the higher the risk the higher the cost to mitigate)
Risk pooling reduces the cost to mitigate risks (eg. pooling demands of several customers reduces the cost to mitigate the risk)
The pooling effect is more pronounced when the pooled risks are very high

Depending on the level of risk and the cost of a mitigation reserve the authors suggest four generic strategic approaches to how to mitigate risks under different circumstances (figure 5).

Figure 5: Risk Mitigation Strategies depending on the Level of Risk and the Cost of Risk Mitigation (Chopra and Sodhi, 2004)

Tailored strategies

Finally the authors show examples of risk mitigation strategies and when to use them. The results are shown in figure 6.

Figure 6: Strategies for Risk Mitigation (Chopra and Sodhi, 2004)

Conclusion

I was surprised by the number of topics the authored covered in a single paper. They really tried to cover the whole risk mitigation process, and they succeeded.

Of course, most of the presented results are not really new and not covered in depth, but they make a excellent basis for strategic, business oriented discussions on supply chain risk management.

One drawback though: I did not really understand why exactly those risk categories (figure 1) were chosen. From my point of view the those are not very intuitive since the categories are not exclusive and thus overlap (eg. a disruption can also lead to a delay. Or why is a information infrastructure breakdown a systems risk and not a disruption?). This makes it harder to use the risk categories efficiently. But the categories could easily be replaced by another framework (eg. by Christopher and Peck, 2004)

Reference:

Chopra, S., & Sodhi, M.S. (2004). Managing Risk To Avoid Supply-Chain Breakdown MIT Sloan Management Review, 46 (1), 53-61

Rate This:

Tags:

Robust Strategies for Mitigating Disruptions

Daniel Dumke — Wed, 25 May 2011 14:51:00 +0000

There are several scientific research centers on supply chain risks in the US (as around the world): The east coast has several researcher on this topic e.g. with Yossi Sheffi from the MIT, in the center with Joseph Fiksel from the Center for Resilience at the Ohio State University and on the west coast Christopher Tang from the UCLA or Hau L. Lee from Stanford University.

I wanted to write about this article, by Christopher and Tang, for some time now.

History teaches us

After a major disruption there are many firms who severely suffer and take a long time to fully recover. On the other side of the spectrum there are a few companies that continue to satisfy their customers demand, and thus use these capabilities to retain and acquire new customers in the process.

Defining robust strategies

Tang defines “robust” strategies as those which fulfill two objectives at once:

Help reducing costs or improve customer satisfaction during times without a disruption,
Support the firm during and after a disruption in continuing operations.

The goal therefore is to create a win-win situation where the investment into the strategy can pay of no matter if a disruption occurs or not.

Cases

Tang lists three cases to highlight some of the few companies that achieved that goal:

Nokia vs. Ericsson
I wrote about this case before, where Nokia and Ericsson were affected by the same disruption at a micro chip plant of a common supplier. Nokia was able to quickly adapt the product design and find another supplier, while Ericsson suffered sever financial consequences leading to the Ericsson / Sony “merger”.
Li and Fung
changed its supply plan in a flash to meet customer demand during a currency crisis.
Dell’s adaptive pricing strategy
to accomodate to supply shortages and still guide and satisfy customer demand.

Robust strategies

Building on these case the author selects and describes nine supply chain strategies that fit above mentioned definition (figure 1).

Figure 1: Robust Supply Chain Strategies (click to enlarge; Tang, 2006)

There are however challenges which have to be considered before implementing the strategies:

Cost versus benefits
These strategies probably make it easier to justify the investments due to their positive effects with or without disruptions, nonetheless there is still a tradeoff which has to be analyzed. Tang suggest to think of the remaining cost surplus as insurance premium.
Strategic fit
Not every strategy may fit for every company. For example Dell’s pricing strategy might be adaptable for airline tickets but not for heavy machinery with a new quote every day.
Proactive execution
Proactive strategies (e.g. rerouting of shipments after a port strike) can be better than reactive strategies (e.g. increasing stocks).

Conclusion

I like this article due to the practice oriented approach taken. But I am missing some clearer description of the ways to generate these strategies. Also, there is no consideration of the completeness of this listing. So I have to assume that it is probably a good starting point. And to my knowledge there are already several studies building on this paper.

As a side note: Nokia also may be a mixed example for a company as role model for good strategies. I would argue that either Nokia was lucky or had a very in-homogenous state of strategy execution: The mentioned disruption happened in 2000, exactly the same time where they should have laid ground for the new smart phone generations which are commonplace now from many manufacturers, but not Nokia.

Reference:

Tang, C. (2006). Robust strategies for mitigating supply chain disruptions International Journal of Logistics Research and Applications, 9 (1), 33-45 DOI: 10.1080/13675560500405584

Rate This:

Tags: