model

Models for Production Planning under Uncertainty

Daniel Dumke — Mon, 19 Mar 2012 16:51:59 +0000

Today will be a one-article-long-excursion in the world of production planning models.
Supply chain management of course should take a high level view of the supply and demand networks, nonetheless there is probably no supply chain which will work without physical products and most even have one or more at their core.

So production planning is a key part of a companies success.
Even in a world where companies focus on their core competencies, and may even outsource all of their production processes, knowledge about how the products are manufactured is still a vital component for understanding the supply chain and the associated risks.

Understanding the models used in production planning supports the awareness of the decision processes, which are applied in production management.

The full article can be downloaded here

Method

In their 2006 article the authors employ a literature review to gather information on which production management models are used.
First, the authors lay the foundations for their work by proposing a classification for models of manufacturing systems (figure 1, for general models; figure 2, for models including uncertainty). For basic model types are distinguished: conceptual-, analytical-, artificial-intelligence-based- and simulation-models

Figure 1: Classification for General Uncertainty Models (Mula et al., 2006)

Figure 2: Classification Production Planning Models under Uncertainty (Mula et al., 2006)

Next, in figure 3 the papers are classified according to the model type used and the decade.

Figure 3: References by Modelling Approach and Year (Mula et al., 2006)

Overall 87 papers are gathered from 1980 until 2004.

Results

Using the categorization suggested above as a frame the authors go through all the papers and add a short description for each of the 87 models. Afterwards the authors conclude:

The analytical modelling approach, in particular stochastic programming was the most frequently encountered. In the case of dynamic programming, few models were found and were mainly theoretical. Most of the analytical models addressed only one type of uncertainty, and assumed a simple structure of the production process. For more complex processes, with many different final products and more than one type of uncertainty, the analytical approach is replaced by methodologies based on artificial intelligence and simulation.
Although many works use simulation approaches to model uncertainty, very few studies exist on the comparative evaluation of the advantages and inconveniences of different simulation languages. With respect to artificial intelligence models, those based on fuzzy set theory represent an attractive tool to aid research in production management. Lastly, conceptual models with different approaches complete the taxonomy.

Conclusion

This article is a great basis to get a short (15 pages) and concise overview on production modelling under uncertainty within the last 22 years.

I was however was surprised to see that the authors did not include clues on how they selected the 87 papers. Or should we really believe that within a 22 year period there were only 87 papers published on production planning?
The authors included also a very short section on supply chain planning models, starting in 1994 with a few papers and I can therefore confirm that there are definitely gaps.

But, as said before, this really is a great overview. And if you are interested in more papers on supply chain planning just have a look here in the blog.

Reference:

Mula, J., Poler, R., García-Sabater, J., & Lario, F. (2006). Models for production planning under uncertainty: A review International Journal of Production Economics, 103 (1), 271-285 DOI: 10.1016/j.ijpe.2005.09.001

Rate This:

Tags:

production

model

Multi-level Supply Chain Design

Daniel Dumke — Mon, 06 Feb 2012 16:48:10 +0000

The quantification of supply chain planning is the next step in the field of supply chain optimization. After operational and logistical aspects have been modeled and optimized, margins for further improvement remain slim.
Based on this premise the paper I review today suggests and tests several alternative multilevel planning approaches to gain further supply chain improvements by optimizing the mid-term supply chain design.

Case

The authors use a case of an agrochemical supply chain to establish their model and methods.
The problem can be stated as follows:

Product X (PX) is a chemical compound used as an active ingredient (AI) in several commercial herbicides. PY is chemically similar to PX, and its uses are nearly identical to those proposed for PX. They are produced by a multinational agrochemicals company.

The authors continue explaining why further optimization is dearly necessary.

A factor that has been putting enormous pressure on the low cost strategy for these products is the price of raw materials. The manufacturing methods are robust and very well established and do not leave any margin for improvement for cost cutting purposes, so the product management team turned to supply chain optimisation as a way of controlling and even reducing costs while improving service levels.

Model

Figure 1 highlights the supply chain structure of this case. The upper part shows a high-level overview, while the lower part displays the structure of the distribution network in the US.

Figure 1: Supply Chain Structure Chemical Industry (Sousa et al., 2008)

The authors employ a two-stage modelling approach to include different aspects of the planning process.

In the first stage we develop a high level planning model with a cyclic time horizon of one year (discretised into twelve months), including all the nodes in the US and worldwide networks as described above.

In the second stage, a detailed operational model is built for each month, with a time resolution of one day to assess the feasibility of the upper level plan at the operational level. […] The US manufacturing sites are described in detail and individual orders are considered.
The outputs are a detailed production and distribution plan for the US network, while accomplishing the export plan established in the first level. The second stage outputs also provide information on how to improve the accuracy of the upper level planning.

The authors then elaborate two mixed-integer linear programs tailored to the demands of the chemical industry. The short-term model is built in a way that environmental variables are used which have been set by the optimization in the mid-term model.
The results therefore can be interpreted as

The objective function of the mid-term and short-term models are the gross profits (NPV). The mid-term model also includes an additional penalty for unmet demand.

Results

For the first / base case figure 2 highlights the percentage of on time delivered products (P3, … P23). Bold numbers are below the 90% target value.

Figure 2: Deliveries on time and in full per Product and per Month for the unaligned Models (Sousa et al., 2008)

It must also be noted that the first stage and the second stage model do not quite fit together. The first stage model consistently projects a higher utilization rate than the second stage (figure 3).

Figure 3: Prediction of Resource Utilization by the first and second stage models (Sousa et al., 2008)

In a second case based on the above mentioned results the capacity of the bottle neck manufacturing sites are relieved. This leads to a slightly higher average percentage of global delivery, but on the other hand also to a lower sales figure for the US market.

Next a multi-level integration of the two different model stages is done. The goal is to use feedback from the second stage model already in the first stage.
Therefore the authors propose the following adjustments to the first stage model:

A capacity correction factor, to adjust selected capacity levels based on learnings from the second stage.
A (reduced) maximum utilization level for certain processes in the stage one model to prevent bottlenecks from happening.
Introduction of a minimum demand coverage by inventory in the first stage model.

After these adjustments the congruence of the two stage models improves and the average on time delivery rises to 97.5 % (figure 4).

Figure 4: Case 3: Increased on time an in full Deliveries compared to Base Case (Sousa et al., 2008)

Conclusion

Multi-level planning is commonly used in research and practice. In businesses very often the planning departments for strategic, mid- and short-term planning are functionally separated. And therefore communication is slowed down.
This article highlights the importance of an integrated planning approach, because if the models are not aligned the end result cannot be optimal.

The authors therefore suggest approaches to adjust the mid-term planning model to the needs of the short-term one. Overall this has quite positive effects on the results of the SC network.

On the other hand the authors neglect to argue in another direction:
The ultimate goal should not be to use more or less subjective adjustment factors and trail-and-error to force the mid-term / first stage model in the right direction, but to integrate supply chain modeling altogether.
Only a fully integrated and comprehensive model can result in real optimization. Of course this would require a whole new, joint planning approach.

Reference:

Sousa, R., Shah, N., & Papageorgiou, L.G. (2008). Supply chain design and multilevel planning—An industrial case Computers and Chemical Engineering, 32, 2643-2663

Rate This:

Tags:

Optimal Design of Supply Chain Networks with uncertain Demand

Daniel Dumke — Mon, 12 Dec 2011 16:02:00 +0000

This week is dedicated to the works on supply chain management from Greek supply chain researchers. Today’s article has been published in the Journal of Management Sciences (Omega) by four researchers from northern Greece and the UK.
After my last reviews which focused more on the conceptual aspects of supply chain risk and management. This paper is again more hands-on in the sense that it describes a mathematical model which integrates supply chain design and uncertain demand and therefore leads to a more robust supply chain design.

Method

The authors propose a mixed-integer linear program to solve a strategic supply chain design problem. Strategic design decisions include:

Where to locate new facilities (be they production, storage, logistics, etc.).

Significant changes to existing facilities, e.g. expansion, contraction or closure.

Sourcing decisions – what suppliers and supply base to use for each facility.

Allocation decisions – e.g., what products should be produced at each production facility; which markets should be served by which warehouses, etc.

Figure 1: Location and possible Locations of Plants and other Facilities (Georgiadis et al., 2011)

Model parameters

To be useful supply chain models usually are limited to a specific supply chain context. In this case the goal is to select an optimal design as well as some tactical / operational parameters.

Figure 1 and 2 describe the location design aspects of the model. The locations of plants and customers are fixed; for the warehouses and distribution centers a set of possible locations is given, and the optimal location has to be selected from the sets.

Figure 2: Locational Limitations to the Supply Chain Design Decisions (Georgiadis et al., 2011)

There are several more constraints implemented, which are concerned with the transportation flows, production resources, safety stocks and capacities. Inventory can be held at different locations, which is solved during the optimization of the model.

The objective is to minimize expected total cost over the planning horizon.

Uncertainty

There are two basic options to integrate uncertainty into a mathematical model:

scenario approach, which discretize the uncertain parameters into a limited number of specified scenarios, or a
probabilistic approach, using stochastic programming.

The authors select the first approach:

In this paper, we adopt a scenario planning approach for handling the uncertainty in time varying product demands. A question that needs to be addressed in this context concerns the generation of the scenarios to be considered. It is, of course, possible to assume that the demand for each product in each customer zone is an independent random parameter. However, more realistically, demands for similar products will tend to be correlated and will ultimately be controlled by a small number of major factors such as economic growth, political stability, competitor actions, and so on.
The complexity of the overall-model is then dependent on the complexity of the basic model (e.g. number of possible connections and locations) and the number of selected scenarios.

Two kind of decisions must be considered in the model: here-and-now decisions (the “really strategic ones”), those have to be selected before any more knowledge about the outcome of the uncertainty can be obtained. The wait-and-see decisions are those which can be altered during a model run. The concept is shown in figure 3.

Figure 3: Types of Decisions in a Strategic Model: ‘here-and-now’ and ‘wait-and-see’ (Georgiadis et al., 2011)

Case study

The authors then set the parameters for the model using an “European wide production and distribution network comprising of three manufacturing plants producing 14 different types of products and located in three different European countries, namely the UK, Spain and Italy” (see figure 1).
The proposed demand volume is given in four scenarios for all customer areas and products. Two cases are compared: one with low safety stock and another one with a general higher safety stock level.
The resulting optimal supply configuration for the high inventory case is shown in figure 4.

Figure 4: Optimal Solution of the Supply Chain Case in the high Safety Stock Setting (Georgiadis et al., 2011)

Conclusion

The authors keep up to their promise and delivered a quite detailed model description and its results. But still, would it be possible to reproduce their results or rebuild their model using this data only? Very unlikely.

Even though the model is detailed. There is still a lot of information missing about the specific parameters used and the interconnections in the model. One major factor in the scientific acceptance and validity of research is the reproducibility of the results. And sadly, that’s one of the reasons, why complex models are still not commonly presented in renowned journals.

In my opinion the only chance to circumvent this obstacle is not only to publish the article, but also the complete model source code and the parameters used.

In the article these omissions are necessary to stay below a certain page limit – the authors already had to distribute the result charts of their case study throughout the paper to have a chance to include the most relevant ones.

But beside these necessary exclusions, I found that some other things would have been interesting to read about.

Demand risk: For a strategic (i.e. long term) model and so many different demand centers, I think only four demand scenarios might be too few to represent reality in a sufficient way. More scenarios could have been included, since the CPU time it took to calculate one optimal solution was quite low (some hundred seconds only).
Other risks: Furthermore it would have been interesting to analyze the effects of other risks in the model, but they were omitted as well.
Overview: I was also missing a short general overview over the given scenarios.
Sensitivity analysis: Lastly, the validity of a model can be further improved by analyzing the sensitivity of the model towards parameter change. The authors did not omit this point, but they choose to test and present only two deviations from their original model parameters, which I think is too little to assess the validity of the model sufficiently.

I think my conclusion can be summarized as follows: It is definitely hard to present a complex supply chain model in a way which sustains the validity and reproducibility of the results. But, since the description of the model is quite elaborate, this paper can still be a great source and foundation for one’s own strategic supply chain model.

Reference:

Georgiadis, M.C., Tsiakis, P., Longinidis, P., & Sofioglou, M.K. (2011). Optimal design of supply chain networks under uncertain transient demand variations Omega, 39 (3), 254-272

Rate This:

Tags:

Matching Product Architecture with Supply Chain Design

Daniel Dumke — Wed, 19 Oct 2011 15:31:00 +0000

This review is about a preprint article which already has been accepted for publication by the “European Journal of Operational Research”. But since there is only a limited space for articles in each issue of the journal, final publication of the article is delayed.
One could now argue in a general note that this behavior also signifies a delay of the progress of supply chain research over all, with all its negative long term effects. Furthermore, in the days of the internet journals should not limit themselves to an artificial (article) limit, but see only the sky or in this case the number of quality publications as their limit.
But in this case I was able to gain access to an early copy.

Product design & supply chain management

I already wrote several times (1, 2, 3, 4) on how product architecture and supply chain design could be integrated, so this is not really a new topic. But it also touches a more integrative approach of supply chain management.

Building from the early models, where a supply chain could easily be defined by a handful of properties, supply chain models nowadays reach a new level of complexity. One of these extensions is the integration of supply chain activities into the product development process, which is supposed to yield (if we trust case studies like IKEA) enormous benefits in every performance aspect of the supply chain and ultimately the company.In the study reviewed Nepal, Monplaisir and Famuyiwa first develop a conceptual and mathematical model to integrate the supply chain design decisions into the product development process. In a second step this model is then tested using two case studies.

Model

The authors suggest an integrated three step process: 1) selection of product architecture, 2) evaluation of potential suppliers, and 3) optimal configuration of supply chain.

The first step includes the development of a generic bill of materials (GBOM) for every possible product architecture scenario. Figure 1 shows the GBOM for an example Product X.

Figure 1: Generic Bill of Materials for an exemplary product (Nepal et al., 2011)

This product design also leads to a similar generic supply chain structure which is shown in figure 2.

Figure 2: Corresponding Generic Supply Chain for Product X (Nepal et al., 2011)

The second step focusses on the identification and evaluation of available the supply chain participants. Necessary information, like production cost, lead-time and compatibility index has to be collected.
The compatibility index itself is developed by the authors as well and contains information on the compatibility of this potential supplier regarding the structural, managerial and financial dimension. Each of the dimensions is then weighted to generate the complete index. Since the compatibility of the future suppliers is hard to judge before the product is even designed, the authors employ fuzzy logic, allowing them to use ranges for each of the compatibility indices instead of fixed numbers.

The last step aims to find the optimal supply chain configuration. This is done for each product architecture scenario by using a linear goal programming model, where total supply chain cost are minimized while maximizing the compatibility between the partners.

Case studies

There are two case studies, one on bulldozer assembly and manufacturing and the other on an automotive climate control system. I will only talk about the first case since I found it more interesting.

For the case the authors exercise the three steps mentioned above. First they design two different product architectures (integral and modular architectures; figure 3) and the corresponding generic supply chain networks (figure 4).

Figure 3: Case Study: Generic Bill of Materials for Bulldozer Manufacturing (Nepal et al., 2011; click to enlarge)

Figure 4: Case Study: Generic Supply Chain Network for Bulldozer Manufacturing (Nepal et al., 2011; click to enlarge)

Lastly, the authors collect the necessary information of the potential suppliers regarding cost, lead-time and ranges for the compatibility index for two different process options:

Two alternatives are considered for each node. If the node is a procurement stage, the first alternative represents the standard supply option (that is, the existing procurement arrangement). The second option represents a consignment option where the supplier is responsible for providing immediate delivery to the bulldozer line. Similarly, for the assembly node, the first option represents the standard manufacturing method while the second option represents an expedited alternative that corresponds to a supplier who has invested in process improvement efforts in order to decrease its supply lead-time.

After solving the model in the next step, the authors conclude:

In the majority of the bulldozer supply chain stages for a modular architectural design, option two has been selected. While it is more expensive than option one, option two has a lower production lead-time and higher compatibility ratings for all stages because its modularity increases the degree of dependency, based on relative costs of inputs, between supply chain nodes.

Compared with base results, where only supply chain costs are considered, this model delivers a solution which performs (slightly) worse in cost (+0,36%), but on the other hand has the potential to deliver a much more balanced and robust solution considering also compatibility issues.

Conclusion

I found the paper to be a great read and I cannot add anything else to the issues the authors already detected and mentioned in their own conclusion:

Risk factors are missing in the model (e.g. random service times)
Additional factors like sustainability and flexibility should be considered
It is very obvious that this is again a supply oriented model, so it would be great to see how distribution network decisions behave when including product design decisions (since the customers might also change depending on the product design decisions)

Reference:

Nepal, B., Monplaisir, L., & Famuyiwa, O. (2011). Matching Product Architecture with Supply Chain Design European Journal of Operational Research : 10.1016/j.ejor.2011.07.041

Rate This:

Tags:

Stochastic Model for Risk Management

Daniel Dumke — Wed, 15 Jun 2011 12:45:00 +0000

Defining a conceptual framework for supply chain risk management can support thinking about risks in supply chains and streamline the decision making process, and therefore improve the current supply chain at hand.
This is similar to a brown-field approach, where gradual changes and risk mitigation strategies are employed onto an existing supply chain. Thus another source for improvement strategies can be a green-field approach, where the supply chain is modeled and optimized to generate new input for real-world optimization.

Model

Goh, Lim and Meng (2007) develop a mathematical model for supply chain optimization. The goal is to optimize facility location and distribution logistics planing in an international setting.

Typically, the firm’s objective is to maximize its global after-tax profit subject to capacity constraints in each plant and demand requirements in each market. The firm thus needs to make an open/shut decision of plants together with the corresponding shipment quantities from such plants to targeted markets taking into account the attendant uncertainties in market demand, volatility in exchange rates, differing country tax rates, and varying import tariffs at different ports of call even within a country.

Solving the model

Two Stage stochastic Model (Goh, Lim and Meng, 2007)

A major part of the model is its objective function, which is displayed in figure 1. In this case the random variables are demand, exchange rate, and import tariff.
Building on the Moreau–Yosida regularization the authors then develop a solution algorithm to solve the problem with profit maximization and risk minimization objectives.

Conclusion

It would be interesting to see this model applied to a real case, as well. But even if the authors presented a real world case there is still a huge gap in this area between the theoretical possibilities and the practical application. Mathematical models for supply chain management have been developed for decades but still mostly the operational models are applied in practice. One reason may be the inherent complexity of the model and the solution algorithms, another might be a managerial reluctance to rely with strategic decisions on models they do not understand. But what’s the alternative? Relying on a gut feel that is more reliable?

Reference:

Goh, M., Lim, J.Y.S., & Meng, F. (2007). A stochastic model for risk management in global supply chain networks European Journal of Operational Research, 182 (1), 164-173 DOI: 10.1016/j.ejor.2006.08.028

Rate This:

Tags:

Monte Carlo Simulation to improve Network Resilience

Daniel Dumke — Wed, 18 May 2011 12:35:00 +0000

Today I want to describe yet another supply chain case study where Monte Carlo simulation is used as decision support for strategic / tactical supply chain decisions.

Case and approach

This time (2005) the authors, Deleris and Erhun from Stanford’s Department of Managements and Engineering, executed a case study with a company (pseudonym: Seltik): “a disguised high-tech company based in the Silicon Valley”.

As a first step the problem was formulated by the managers, who were concerned if their strategic network design was adequate for the risks their company faced.
To answer this question the authors use two building blocks: a network flow model which in turn feeds into a Monte Carlo simulation.

Figure 1: Risk Management Approach (Deleris and Erhun, 2005)

Risk management and network definition

As a first step possible risks were collected and evaluated following the process depicted in figure 1. The “loss of volume” was selected as performance measure and to measure the vulnerability of the supply chain.

In the next step was to model and visualize the supply chain network. In this case the authors focussed only on the most profitable products. The result has been summarized in figure 2.

Figure 2: Seltik’s Simplified Supply Chain Model (Deleris and Erhun, 2005)

Flow model and simulation

The goal of the flow model is to determine the effects of a supply chain disruption to one of the nodes or links on the performance measure.

Consider for instance a single-product network composed of two paths A — d — C and B — e — C, where capital letters denote nodes and bold lower case letters denote arcs. Each path transports 50% of the product volume. Assume that A is not operating for instance due to a strike, then the loss of volume is 50%. If in addition, the arc d is broken, for instance due to a storm, then the loss of volume is unaffected.

In this model a node / link can only be open or closed. The result of this analysis is a probability distribution of the volume loss. The process is depicted in figure 3.

Figure 3: Input Information used by the Supply Network Risk Assessment Tool (Deleris and Erhun, 2005)

However the flow model only results in a deterministic analysis of the network performance under risk. Monte-Carlo simulation provides the means to generate a “risk independent” statement on the resilience of the network towards a wide range of risks. The following risk events were included within the analysis:

the possibility of employee strikes,
the shortage of components,
severe political instability in the various regions, and
disruptions caused by hurricanes.

The simulation was performed using Excel and @Risk (Excel Addon) using a simplified model, with a focus on the critical (non-commodity) raw materials. The results show that in 75% of the scenarios no loss occurs, for the other 25% figure 4 depicts the contingent probabilities of the corresponding loss.

Figure 4: Probability Density Function for Quarterly Losses (Deleris and Erhun, 2005)

Conclusion

The model presents a possibility of how to include Monte-Carlo simulation into a decision support system. Although the paper is not detailed enough to directly reproduce the model. Nonetheless, the authors also provided the risk data which was estimated during the case.
There are a few drawbacks to using this model directly.

The model used here is very basic and does not consider partial disruptions: links / nodes are either open or closed, nothing in between.
Even if using Monte-Carlo Simulation to make the risks / environment dynamic, the network is still static. So there is no reaction to any disruption, like shifting sourced material to the more stable supplier.

Reference:

Deleris, L.A., & Erhun, F. (2005). Risk Management in Supply Networks using Monte-Carlo Simulation Proceedings of the 2005 Winter Simulation Conference, 1643-1649

Rate This:

Tags:

model

monte carlo

disruptions

Supply Chains and Fuzzy Demand

Daniel Dumke — Wed, 04 May 2011 15:33:00 +0000

Risk in supply chains can be included in several different ways into the decision making process.

No Risk

A statement in many supply chain models is that some/most/all parameters of the model are fixed (e.g. fixed demand, zero probability of a hurricane).

The result is, if the real value of this parameter diverges from the assumptions, the results of the model will be flawed to a certain degree (up to completely unusable).

Distributions

Another allegedly more complete way of including risks into modeling is to assign specific probabilities or even stochastic distributions the model’s parameters. In this way using a normal distribution, would represent the uncertainty about the future values of the parameter. Nonetheless, this approach has theoretical and practical flaws as well. From a practical standpoint, the time needed for solving the resulting model (especially with realistic model sizes) increases dramatically and may even be insolvable. And from a theoretical point of view, the assigned distributions may also prove to be wrong in reality.

Fuzzy Sets

Fuzzy logic is another way of introducing risk considerations into a model. This article is about the theoretical application of this logic to a supply chain problem.

Fuzzy sets are a construct which can be used to a limited amount of uncertainty into a system’s model.

A fuzzy demand variable, as an example, could have four possible manifestations of (14, 16, 18, 19). This can be interpreted as the demand can have one of those values without further specification of the individual probabilities.

Model and Results

Wen and Iwamura (2008) developed a facility location allocation model where the demand parameter has been defined as a fuzzy variable.

Since the demand in this case is not deterministic anymore, decisions have been based on an expected value.

The authors used the Hurwicz criterion, a decision function based on the weighted average of the worst and best outcome of a given decision, as the objective function to evaluate the “best” location-allocation.

They also suggest and test an solution heuristic. A possible solution of this location problem is shown in figure 1.

Figure 1: Exemplary Results: Location of Customers (dot) and Facilities (diamond) (Wen and Iwamura, 2008)

Conclusion

One advantage of the concept of fuzzy sets is that it may be easier for an analyst or expert to specify a limited number of possible demand values than fill the parameters an appropriate distribution function.

The disadvantages lie in the further assumptions of the fuzzy sets.

The fuzzy variables are defined to be trapezoidal, ie. probabilities are assigned to the possible manifestations eg. (14, 16, 18, 19) in such way that the probability for the extreme values are lower than the center values. And this is basically the representation of a discrete distribution function.

There is no evidence for fuzzy logic as better representing the uncertainty in a real supply chain system

The data shows that the computational efficiency does not seem to be very good

Furthermore, fuzzy logic does not allow to represent extreme and unlikely events in a adequate manner. In the case of demand, there may be the possibility of extreme high or low demand which could lead to a change in the optimal locations.

Reference:

Wen, M., & Iwamura, K. (2008). Fuzzy facility location-allocation problem under the Hurwicz criterion European Journal of Operational Research, 184 (2), 627-635 DOI: 10.1016/j.ejor.2006.11.029

Rate This:

Tags:

Risk, Information and Incentives in Telecom Supply Chains

Daniel Dumke — Wed, 09 Feb 2011 16:26:00 +0000

Supply chains risks can also be analyzed in a specific industry context and this is exactly what Agrell et al. (2004) did with telecom supply chains. They used a three tier SC (2nd tier supplier, EMS, OEM) to include the selection, coordination and motivation of independently operating suppliers in the model.

In terms of risk handling and sharing the telco industry is to some extent unique as well; there are several possible complications, like

differing business logic in the different stages,
individual relations between revenue and product life cycles
design of products and processes and
a lack of effective incentive structure to induce global supply chain optimization promotes the opportunistic and myopic behavior of the chain firms.

Figure 1: Historical Development within the Telco Industry (Agrell et al., 2004)

Telco Industry

The telco industry can be described using the following dimensions.

Figure 2: Diverging Business Logic in the Telco Industry (Agrell et al., 2004)

Demand uncertainty
- Level
- Timing
Changing Roles
- Outsourcing
- Unclear Interfaces
Growth vs. Consolidation
- Shifting power balance due to consolidation at supplier level
Heterogeneous Business Logic (see Figure 2)
- The cost and revenue functions are different

Model

The authors developed a three stage model with the 2nd tier supplier (for components), the EMS (Electronics Manufacturing Services Provider) and the OEM (Original Equipment Manufactur) in focus (gray box in Figure 3).

Figure 3: Base Supply Chain Model (Agrell et al., 2004)

The experiments with the model take place using a two period time horizon. The OEM tries to maximize the profit over this period.
Knowledge about the system is explicitly divided between the participants: Common knowledge: current demand, Supplier: its cost function and investment opportunities.
Information can be shared: The OEM can send a forecast to its suppliers and it sends order quantities for the first and second period to its suppliers.
The following alternative scenarios are then compared:

Centralized model
OEM coordinates with asymmetric information
Coordination by EMS

Results

The differences between the scenarios are quite low, nonetheless the following conclusions can be drawn:
The performance of the supply chain as a whole may be undermined by the shifting positions of bargaining strength (towards the suppliers).
And the model shows furthermore that simple price-quantity-only coordination (without long-term contracts) as some shortcomings which leads to reduced performance as well.

Conclusion

One question the article did not answer: Why specialize on the Telco industry? The above mentioned description of the industry might fit on others as well, so it would have been more straight forward to develop the model based on the abstract scenario description and use the telco industry as a case study.
Contrary to the description, the model is basically a four-tier model. Since also the customer of the OEM is taken into account as the final source of demand.
But in my view what stands out is not the number of tiers in the model, but the very explicit modeling of the knowledge distribution. Many papers only take a very brief detour on the knowledge that each participant has, or leave it out entirely. But not only for the quality of the model, but also its reproducibility it is important to take the information distribution into account.

Reference:

Agrell, P., Lindroth, R., & Norrman, A. (2004). Risk, information and incentives in telecom supply chains International Journal of Production Economics, 90 (1), 1-16 DOI: 10.1016/S0925-5273(02)00471-1

Rate This:

Tags:

information

model

performance

Level of Detail in a Simulation Model

Daniel Dumke — Mon, 29 Nov 2010 14:48:00 +0000

After the last more general entries on managers perception of risk and measuring SC performance I wanted to make a detour back to the basics.
Simulation is one of the tools, which can be used for analyzing supply chain dynamics, optimization and to support corporate decision making.
One major question when starting a supply chain model has always been what level of detail should you choose? Someone could start with a single worker in an agent based simulation model and continue with the machine he is operating, but when the goal is to gain insights about the interaction between your company and your closest competitors this might be too much information.
To get a better grasp on this topic I read “The impact of different levels of detail in manufacturing systems simulation models” by J.F. Persson (Translation).

Experiment / Results

To test the importance of the level of detail in a simulation model Persson designed an experiment: He modeled the same process chain with three different levels of detail:

high level of detail
some aggregation
only main processes

After simulating this process and comparing the output measured in capacity utilization, inventory levels and blocked time (for bottleneck detection), he concluded that significant differences between the models have to be acknowledged (see figure).

Capacity at different Stages within the Models (Persson 2002)

Conclusion

So what are the important lessons for someone planning to develop and use a supply chain model:

Focus
Know your exact goals and focus on achieving them. If you want to analyze competitive interaction between companies, it might be a good idea to have a single company as your base entity
Minimization

The level of detail should be kept at a minimum, keeping the model simple. The reason is that (i) a simple model can provide a result and (ii) an unimportant detail in the model is hard to remove despite the negative impacts on simulation execution time.
Validation
It is very important to have constant validation for the model during the development process and final approval tests to prevent adverse effects from a wrong level of detail as well as other mistakes.

Reference:

Persson, J. F. (2002). The impact of different levels of detail in manufacturing systems simulation models Robotics and Computer-Integrated Manufacturing, 18 (3-4), 319-325 DOI: 10.1016/S0736-5845(02)00024-8

Rate This:

Tags:

model

Simulation