2-Sales and Operations Planning The Supply Chain Pillar
2-Sales and Operations Planning The Supply Chain Pillar
2-Sales and Operations Planning The Supply Chain Pillar
To cite this article: R. Affonso , F. Marcotte & B. Grabot (2008) Sales and operations
planning: the supply chain pillar, Production Planning and Control, 19:2, 132-141, DOI:
10.1080/09537280801896144
In the present ever-changing environment, sales and operations planning (S&OP) is a key process for providing
visibility to the enterprise. Besides, it supports a transversal decision process, which co-ordinates different
functions either in the company or between companies, in a supply chain (SC) environment. In the literature,
S&OP models are mainly focusing on sales, production and inventory. This paper proposes a wider S&OP model
built with three levels (sales, operations and supply). It provides a better support for integration inside the
company, but also for integration of the company in the SC. Some simulation results are presented, describing
the S&OP propagation along a SC, and the related collaboration requirements to be satisfied between the
networked companies.
Keywords: sales and operations planning; supply chain; collaboration; planning optimisation
The understanding of co-ordination mechanisms integration intra- and inter-companies, while keeping
can help managers in the decision-making process to their decision autonomy among the SC.
select the most appropriated action from a set of Sales and operations planning is a major planning
alternative solutions. Integration mechanisms can process that has an important integration power,
help them to define to what extent such action/ connecting different functions of the company
interventions should pass through organisational (marketing, financial, production, etc.) with various
boundaries between functions and between companies points of view, objectives and constraints.
(Romano 2003). The S&OP process supports vertical integration, in
Several studies have been performed providing relating strategic and financial plans to operational
different ways to better co-ordinate and integrate plans. It also supports integration between companies
companies across the SC. Chung (2005) proposes in the SC, simply linking the company’s commercial
collaborative planning, forecasting and replenishment department with the customer purchase service,
(CPFR) to integrate companies, developing and and the company’s purchasing department with the
monitoring business plans and forecasting between commercial department of the suppliers (see Figure 1).
partners. CPFR is applied in a context where the So the S&OP process (see Genin 2000) remains an
customer takes the product directly from the inventory integration pillar in the SCM.
store. It aims at a more precision in sales forecasts For these reasons, we have been interested in
and inventory reduction, but other members of the SC analysing a supply chain co-ordination and integration
are not directly involved in the collaborative through the S&OP. Such a study also provides
relationship. elements to structure the overall collaboration process
Eng (2005) suggests mobile SCM to extend both in the SC.
intra- and inter-firm business systems by enabling SC So this paper suggests an S&OP model that takes
participants to carry out business activities such as into account the influence between various functions in
online transactions performance, sharing and exchange a company, and the relations between companies in
of up-to-date information, providing customer service a SC. In order to illustrate the behaviour of the S&OP
on demand, managing logistics, transportation and process along the SC, some simulations have been
inventory levels. performed to analyse the collaboration requirements,
ERP (Kelle 2005), Portals (Carlsson 2004) and depending on a SC typology.
Internet (Garcia-Dastugue 2003) are also used as tools The paper is organised as follows: we first present
to co-ordinate companies in the SC. They manage an S&OP model. Then, a SC structure and typology
flows of business information, process and transac- is described, aiming at testing the S&OP behaviour
tions. These proposals are used in a centralised across the chain. In a third part are presented some
management of the SC. These research works mainly simulation aspects. Finally, we discuss some prelimin-
bring solutions on how to support collaboration, but ary results and perspectives of work.
provide less support on how to structure collaboration
process.
On the other hand, interviews of SC managers
3. S&OP: model proposal
show that SC actors usually have a poor visibility on
the internal process of their partner. Therefore, the 3.1. Model overview
supply chain management is often reduced to a peer- In the literature, most of the S&OP models focus on
to-peer relationship. Besides, the economical and the relations between commercial requirements (sales
juridical boundaries between the companies enable forecasts) and production resources (mainly capacity
managers to keep the autonomy of decision among oriented), where the differences are managed through
the SC. inventory projections. For instance, Table 1 illustrates
Some authors (Thomas 2000, Genin 2000) consider how the inventory at the end of a period is calculated
that the sales and operations planning (S&OP) is on the base of the expected sales, production plan and
a relevant process to provide co-ordination and inventory at the previous period.
S&OP in the SC
Sales Purchase Sales Purchase Sales
Company 3 Company 2 Company 1
Table 1. Classical S&OP model (Vollman 1997). Adjusted capacity: Represents the capacity taken into
account for a period, after an adjustment decision
Sales and operations planning based on the available flexibility.
M Mþ1 Mþ2 ... Operations plan: Represents the expected production
for a period, based on the adjusted capacity.
Sales plan 140 135 150
Production plan 145 145 145 Inventory: Represents the inventory of the previous
Inventory 15 25 20 period plus the inventory variation.
Initial inventory 10
Monthly cost: Represents the inventory cost, plus the
production cost and capacity cost (average capacity
The S&OP horizon covers at least the overall lead costs plus adjustment costs).
time (supply and manufacturing). The figures are Total cost: Represents the cumulated cost along the
distributed monthly (monthly period) and the units S&OP horizon.
are product families.
In this classical S&OP model, the main adjustment
Supply level
parameters are inventory and capacity. However,
the manufacturing and supplying constraints are Material inventory: Represents the inventory available
wider and should be taken into account for an efficient for manufacturing (raw material and purchased parts).
co-ordination of the enterprise functions. Supply plan: Represents the quantity to be supplied,
Also, among a supply chain, the relations between based on the production plan and on the material
companies and the constraints regarding the overall inventory.
process should be taken into account. For that The relations between these levels can be described
purpose, we propose an S&OP model which specifi- as follows: the sales plan will be built from the
cally describes the relations between sales, operations forecasts, and will be the input for the operations
and supply. Then, considering that the supplying level. At the supply level, the supply plan will be built
requirements for one company become the sales from the operations plan. Furthermore, to calculate
requirements for the upstream company in the SC, the S&OP, the lead time between each level must
it becomes possible to analyse the relations between be taken into account. It means that the supply plan
the companies, together with the constraints will be shifted from the operations plan of the
propagation along the SC, at the level of the S&OP. production lead time (see Figure 2).
The proposed S&OP model, based on three levels, These relations between plans are easy to integrate
is described below. in the model, as long as the lead time between two
stages is a multiple of the period (month). In the case
Sales level of a one month lead time, the quantity to be supplied
will be directly shifted one month period from the
Forecasts: Represents the gross value of the sales operations plan.
forecasts, in terms of product family per period, However, if the lead time is not a multiple of the
provided by the sales department. period value, the relation between plans is more
Variation: Represents, per period, the possible varia- complicated. The quantity to be produced from the
tion of the forecast, which characterises the forecast operations plan will be split between two periods.
reliability. For example, the supply requirements for one period
Sales plan: Represents the sales forecasts finally will correspond to a mix of the production
considered for the S&OP process.
Operations level
Forecasted workload: Represents the work load,
induced from the sales plan for each period.
Available capacity: Represents the amount of capacity
available for a period.
Flexibility: Represents the amount of extra capacity
(extra hours, subcontracting) for a period (addition of
workers, subcontracting, new equipment, etc.). Figure 2. Shifting between the S&OP.
Production Planning & Control 135
capacity is modified, and inventory cost). The labour industry, computer manufacturer, and so on. Most of
costs are fixed for the nominal capacity. The nominal the product value is added at the last stage.
capacity is the average capacity established to balance The second structure illustrates a distribution
the SC regarding its market perspectives. The adjust- network with the main company at the first stage of
ment costs are dependant on the labour cost (capacity the SC.
cost), on the amount of adjustment and on the S&OP Considering these two types of costs structure
horizon zone (adjustments performed at short term are (see Figure 8), we analyse different cases of the
more expensive than early adjustments). The more S&OP propagation, related to a demand forecast
important the adjustment is, the more important the perturbation.
related over cost is.
Then, we have established different costs ranges
related to the labour adjustment percentage and to the 5. Simulation
different S&OP horizon zones (see Table 3). Thus, it is 5.1. Simulation scenario
possible to represent in the model the difficulty that After defining the S&OP model and the SC structure,
companies have to manage demand variations, we have performed some simple simulations to analyse
depending on the visibility they have and on the the model, and the benefits of adopting collaborative
amplitude of the variation. approaches in the SC.
At each level of the SC, the cost of each product A preliminary simulation has been realised to
(wood, leg, top and table) will be based on the labour adjust the model parameters. In this context, we have
costs, adjustment costs and on the raw material costs. verified that (as mentioned in Section 3) in the case
The table costs are based on assembly costs, adjust- where the lead times are not a multiple of the period
ment costs and on legs and top costs. The costs of the value, some demand forecast perturbations were
top of the table will be based on top production costs, shared between two different periods. This effect
adjustment costs and on wood costs, etc. In the hides some capacity problems in the S&OP calculation
simulations, the inventory costs are considered as process. This is why we have decided to consider a one-
15% of the stored product cost. Regarding the overall month lead time only for the production steps
costs structure, there are different types of situations in the SC.
(see Figure 7).
In the first situation, the labour costs are higher in
the last company of the supply chain. Thus, the final
product cost is mainly added in the last company of the
SC. This type of cost structure is denoted here as
downstream added value.
In the second situation, the labour costs are higher
in the first steps of the SC. This type of situation will be
called upstream added value.
The first structure corresponds to a classical SC
product transformation, where the last company is
often the leading one: car manufacturer, aircraft
Figure 7. Cost structure.
30
Adjustment Fixed Flexible Free 20
10
Cap.550% þ30% þ30% þ30% 0
50%5Cap.560% þ10% þ10% þ10%
Wood Top/Leg Table
60%5Cap.580% þ10% þ10% 0
80%5Cap.595% þ10% 0 0 Companies
95%5Cap.5105% 0 0 0
105%5Cap.5120% þ40% 0 0
120%5Cap.5140% þ40% þ25% 0 Scenario 1 Scenario 2
Cap.4140% þ60% þ35% þ25%
Figure 8. Graphic of the added value.
138 R. Affonso et al.
In order to analyse constraints propagation among (see Figure 9), it is more interesting to optimise the
the SC, we have considered one month on the horizon, S&OP in the downstream of the SC (Table 4). In this
where the sales forecast is significantly higher than the case, the collaboration does not bring better solutions
average forecast along the S&OP horizon. It means for the global performance.
that, in our model, the sales forecasts are constant, If the impulse, still in the free zone at the level of
except for one month, where there is an increase of the the table, is closer to the flexible zone on the horizon,
forecast (50%), always in the free zone. We call this so that the impact appears in the fixed zone or even
forecast increase of impulse. in the flexible zone for the Wood Company (see
Based on these assumptions, two parameters were Figure 10), the capacity adjustment is more expensive
adjusted: the month where the forecast perturbation if it goes over 5%.
appears at the Table Company level, and the cost In this situation, the inventory costs (weak at the
of capacity adjustment. This last parameter allows beginning of the SC) lead the Wood Company to
testing different levels of flexibility in the SC. More smooth the workload. The limits are reached when the
flexible is the company, less expensive are the adjust- wood company has not enough horizon to smooth the
ment costs. workload.
Depending on the impulse month for the down-
stream company (Table Company), the impact of this
impulse may appear in the flexible or even in the fixed 6.2. Scenario 2
zone for the upstream company, which is the Wood
In this situation (see Figure 11), we expected a better
Company.
SC performance with collaboration, but the conclusion
To analyse the benefits of collaboration in the SC
depends on the month when the impulsion occurs
at the S&OP level, two kinds of decision among the
(see Table 5).
SC were tested.
If the forecasted perturbation is far enough on the
. In the first case, each company makes and horizon (see Figure 12), the adjustments for all
optimises its own S&OP and transmits the companies take place in a zone where the costs are
results to its supplier in the SC. Such a local
optimisation is somehow against the SC
philosophy.
Scenario 1 Cost Graphic
. In the second case, the aim is to optimise the
30
local costs (sum of all the costs for all
Cost (€)
Scenario 1
10
5
Sales forecast Cost without Cost with
variation month collaboration (E) collaboration (E)
0
Wood Top/Leg Table April 6652 4722
May 4493 1720
Companies June 2246 622
Inventory Cost Labour Cost
Figure 11. Scenario 2 cost graphic. Obviously, in this case, collaboration appears much
more interesting.
Table 5. Scenario 2 – simulation results.
Scenario 2
7. Conclusion
S&OP is considered as a very important tool to
Sales forecast Cost without Cost with co-ordinate and integrate companies in the SC.
variation month collaboration (E) collaboration (E)
However, when all information and parameters
April 5660 4722 needed to implement this integration are taken into
May 1672 1720 account, it becomes a complex subject, with several
June 419 622
rules to be clarified.
The proposed model was built with three levels
(sales, operations and supply). The purpose was to
formalise the upstream relations (supply plans) and
the downstream relations (sales plans) for a company
in the SC, at the level of the S&OP.
The simulation results have several limits, the main
one being that the simulation was performed on a
specific example (even if we have tried to be as generic
as possible when defining the test case). Building the
performance indicators we only focussed on costs,
and a simple cost structure was used. Customer service
was not addressed, while it is an important purpose
of collaboration in the SC.
Figure 12. Example of S&OP propagation in a SC (2). For the input, only a local forecast increase was
considered as ‘constraint’. We did not analyse the
collaboration in the SC to release other constraints
such as over load, down load, missing parts, specific
less important (free zone). Then, they lead the Wood customer requests, etc.
Company to smooth its own production and increase In our simulation, the companies have no capacity
the inventory (the inventory costs are always cheaper limitation, but only extra costs for capacity adjust-
upstream). ments. Finally, the optimisation during the simulation
If the forecast impulse for the assembly company was made manually. In spite of these limitations, this
is still in the free zone, but closer on the horizon, the work allows us to make recommendations to support
smoothing for the Wood Company becomes much SC control at the level of the S&OP.
more expensive (impact for the Wood Company in Considering the S&OP process as the SC pillar is
the fixed zone, see Figure 10). Then, the collaboration relevant as long as the three main sectors of the
provides better opportunities. company are taken into account: sales, operations and
All these simulations were performed with the supply. Then, it is important to define the rules to
flexibility costs shown in Table 3. In Table 6 we manage the time-phased requirements.
present the results of a simulation with no free The identification of specific horizon zones, such
adjustment possible in the SC, in order to illustrate a as time fences (frozen, slushy, liquid) usually used for
poorly flexible SC (important adjustment costs). the MPS (Wollman 1997), is also very important.
140 R. Affonso et al.
Eng, T.Y., 2005. Mobile supply chain management: Romano, P., 2003. Co-ordination and integration mechan-
challenges for implementation. Technovation, 26, isms to manage logistics processes across supply networks.
682–686. Journal of purchasing and supply management, 9 (3),
Genin, P., 2000. Le Plan Industriel et Commercial: Recherche 119–134.
de Pratiques Optimales du Processus d’Elaboration. Master Sahin, F. and Powell, R.E., 2005. Information sharing and
Rapport of Marne-La-Vallée University/National School co-ordination in make-to-order supply chains. Journal of
of Ponts and Chaussées. operations management, 23 (6), 579–598.
Mentzer, J.T., Min, S., and Zacharia, Z.G., 2000. The nature Stock, G.N., Greis, N.P., and Kasarda, J.D., 1998. Logistics,
of interfirm partnering in supply chain management. strategy and structure. A conceptual framework.
Journal of retailing, 76 (4), 549–568. International journal of operations & production manage-
Kelle, P. and Akbulut, A., 2005. The role of ERP tools in the ment, 18 (1), 37–52.
supply chain information sharing, cooperation, and cost Thomas A. and Lamouri S., 2000. The new problem
optimisation. International journal of production economics, with sales, inventories and operations planning in a
93/94, 41–52. supply chain environment. In: Proceedings of the
Lee, H.L. and Ng, S.M., 1997. Introduction to the special International Society for Optical Engineering on
issue on global supply chain management. Production and Intelligent Systems in Design and Manufacturing. Boston,
operations management, 6 (3), 191–192. USA, 321–329.
Malone, T.W., 1987. Modeling co-ordination in organisa- Vollman, T.E., Berry W.L. and Whybark, D.C., 1997.
tions and markets. Management science, 33 (10), Manufacturing planning & control systems. 4th ed. New
1317–1332. York, NY: McGraw-Hill.