Hard Problems and Soft Information
Coen Suurmond
RBK Group, Keulenstraat 18, 7418 ET Deventer, The Netherlands
csuurmond@rbk.nl
Keywords: Business Processes, Process Logic, Organisational Semiotics, Sales & Operations Planning.
Abstract: Finished product planning is a key business process for companies. It is about finding the balance between
service levels and cost, and is therefore critical for the success of the company. In this paper the structure of
the problem will be analysed and compared with literature about sales & operations planning as well as ERP
solutions. In the analysis general process logic will be contrasted with idiosyncratic characteristics of the
individual company. The use of different kinds of information will be discussed, in combination with the
formal sign system of the computer and the social sign system of human communication.
1 INTRODUCTION
Finished product planning is a key business process
for companies. This planning decouples demand from
production and it is a highly determining factor both
for market volume and for profit margins. On the
market side customer service level and lead time are
key for market share while on the production side it
is crucial to keep down variable costs. In this paper a
specific product group (fresh meat) for a specific
market (multi-store retail chains) will be used to
analyse the business processes and information flows
that are in play, and it will be examined how this
problem is covered in literature and by ERP software.
In this paper the paired concepts of process logic
/ idiosyncrasy will play an important part. Process
logic will mark the necessary general structure
underlying the actual business processes of the
individual company. This structure will always be
present in every business in a particular market,
because certain structures are inherent to and
inevitable in operating with these products on those
markets. Alongside this there are the idiosyncratic
characteristics of the individual company. Companies
after all do differ, even though they operate with the
same products in the same market segment. The
individuality of the company is the foundation of the
existence of any company on the market. As John
Kay pointed out, the distinctive capabilities of the
company are what distinguishes the company from its
competitors and that form the foundation for the
success of the company. And, again as pointed out by
John Kay: “A firm can achieve added value only on
the basis of some distinctive capability – some feature
of its relationships which other firms lack, and cannot
readily reproduce” (Kay, 1993, p. 64).
The analysis of business processes in terms of
process logic and idiosyncratic characteristics has
two aims. Firstly, it is a method to map business
processes in a way that helps external consultants
(specialists in general patterns) and internal
employees (specialists in specific details) to come to
a mutual understanding and a common basis.
Secondly, it is about the awareness of the intangible,
the understanding that not everything can be reduced
to schemas and fixed rules. Information systems
should not be a goal onto themselves but rather serve
to improve the market position of the company, which
they must do by adequately supporting the business
processes. The process logic provides insight into the
general structures, while the idiosyncratic
characteristics provide insight into the way in which
the processes actually happen within the company
(including the often blurred boundaries between
processes). This approach should also help in getting
a feel for the distinctive capabilities of the company,
competitive strengths that must be preserved and
possibly enhanced in developing a new business
information system.
Besides the paired concepts of process logic /
idiosyncratic characteristics the analysis of the nature
of the information in business processes with relation
to formal and social sign systems used forms a second
pillar of this analysis. Computer systems are formal
sign systems, highly capable in processing declarative
information, but they have trouble with vague
32
Suurmond C.
Hard Problems and Soft Information.
DOI: 10.5220/0006221900320038
In Proceedings of the Sixth International Symposium on Business Modeling and Software Design (BMSD 2016), pages 32-38
ISBN: 978-989-758-190-8
Copyright
c
2016 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
boundaries, “unclean” categorisation and weighing
heterogeneous norms that cannot be fulfilled
simultaneously against each other (delivery in full but
slightly delayed or delivery on time but slightly less
than ordered? Delivery should be at the right time in
the right quantity with the right quality, which ‘right’
might be relaxed in which context?). Social sign
systems are much better in dealing with meaning in
context, modalities, and intentions (discussed in
Suurmond, 2015). In the creation of a business
information system awareness of the nature of the
information and the conscious choice for the right
sign systems is critically important for the
effectiveness and efficiency of the business
processes. Of the coordination mechanisms identified
by Mintzberg direct supervision and mutual
coordination are much more based on informal sign
systems, while the application of formal sign systems
presupposes standardisation (Mintzberg, 1979). The
effectiveness and efficiency of business processes is
of course dependent on correct, timely and complete
information (Starreveld, 1963), but also dependent on
the degree to which the available information is
relevant and accessible (Grice, 1989).
The paper consists of six sections. After a short
introduction to the problem area, the business
processes involved are analysed in terms of process
logic and idiosyncrasy. This is followed by an
analysis of the nature of the information used in the
business processes, and the role of formal and social
sign systems in conveying the information. Literature
about sales and operations planning is discussed in
section five, and the papers ends with a recapitulation
and conclusions.
2 PROBLEM AREA
This paper is concerned with the finished product
planning for the production unit for pre-packaged
meat in retail chains. A typical production unit
produces some 100 to 200 different fresh pre-
packaged meat products on a daily basis. Incoming
shop orders need to be made available ready-to-ship
on the loading dock for transport, delivery reliability
must be above 99.7%. The products have an internal
shelf life of at most two days. Everything is produced
from fresh ingredients each day. And, of course,
waste and production costs must be kept to a
minimum.
Many products have a fairly stable demand
pattern. Demand is however highly irregular in case
of promotions, product introductions (which do not
have a demand history) and a number of articles that
are weather-dependent. Products which have been
part of a promotion in the past weeks and seasonal
products also have demand irregularities.
Planning is generally done in a large Excel in
which the order history over the past weeks for each
distribution timeslot is recorded along with the
demand prognosis for promotions and in which the
planner records the amounts per product per day that
need to be produced. This last list is processed further
within the production planning to create production
orders as well as lists for the resource, man-hour and
line-hour needs. Each business within this sector has
developed its own particular solutions over time, the
common characteristic is the use of Excel with order
history as input and production lists as output. The
expected demand for promotions is mostly
determined in a separate process and then made
available to the planner.
3 PROCESS LOGIC &
IDIOSYNCRASY
3.1 Process Logic
Every company that produces for a market in which
the lead time between order time and delivery time is
less than the time necessary to produce the goods will
work with a (semi) finished product stock. This stock
must be sufficient to fulfil the orders within the
delivery reliability requirements in this market.
Further, the stock has to be as low as possible because
of warehouse costs and to minimise waste. Finally,
production will have requirements regarding the
frequency and size of the production batches. All of
this leads to the following three processes that will
always be found (but, more often than not, implicitly
rather than explicitly):
1. Determining expected demand
2. Determining the target stock level
3. Determining the production output
The distribution pattern defines loading times for
groups of orders, before the scheduled loading time
the orders must be picked and made available at the
loading docks. Therefore, distribution and production
is organised in timeslots, in each distribution timeslot
a set of orders is picked, and in each production
timeslot a set of products is produced. Shops have a
timeslot for ordering. Expected demand is specified
per distribution timeslot, which would correspond
with one or more ordering timeslots. The end times of
ordering timeslots and distribution timeslots are
Hard Problems and Soft Information
33
fixed, the end time of production timeslots is more
flexible.
Given the daily delivery to the shops in
combination with the internal shelf life of two days at
most, the full product range will be produced and
distributed in a 24 hour cycle. For each day and each
timeslot expected demand will be estimated, target
stock levels will be set and production output levels
will be set. The planning moments are:
1. After each ordering timeslot
2. Just before the production day
3. Just before the production week
4. A few weeks before the production week
5. About four to six weeks before the production
week
After each ordering timeslot the planning is checked
and adjusted where necessary (and possible), one day
in advance production orders are generated and fresh
raw materials are ordered, the week before the
production week production schedules are set and
suppliers will be informed about expected demand,
and the same applies for the planning moment a few
weeks before the production week (same information,
less certainty. In the first planning moment about four
to six weeks in advance of production promotions are
planned (because of the increased quantities it is
important to have agreements in place with suppliers
regarding price and expected demand of raw
materials).
3.2 Idiosyncrasies in Individual
Companies
Although the processes described above in the
process logic can be found in each company in this
market, in organisation and actual execution a great
variety exists. Often, boundaries between the
processes are blurred, both organisationally and in the
Excels used for planning. Much of the knowledge and
information is personal and non-coded (Boisot,
1998). Many problems are either ‘spirited away’ by
some creative and experienced old hands, or solved in
informal communication. Sometimes, this is a good
thing, because the problem solving capabilities of the
company are much greater than one would expect
from studying organisation charts and documents
about process flows. Sometimes, it is good but too
vulnerable because of the dependencies on one or two
key figures in the organisation. Sometimes, it is bad
because problems are not really solved, only moved
out of perception.
4 NATURE OF INFORAMTION &
SIGN SYSTEMS
4.1 Information for Determining
Expected Demand
In the process ‘determining expected demand’
(forecasting) the single information product appears
to be a list with the expected shop orders grouped by
timeslot and by saleable item, and the primary source
of information is demand history. However, although
extrapolation from history to expected demand may
lead to a convenient list of hard data, it does not tell
the whole story. Firstly, demand history is never the
only source of information. In case of promotions and
product introductions no relevant and reliable
demand history is available (although, the history of
promotions and product introduction might offer
useful indications), and demand for some products
might be dependent on future conditions (weather) or
situations (events, bank holidays). Other information
sources are needed, and the information from these
sources must be interpreted in context. The
interpretation of the weather forecast, especially in
case of a possible sudden change of the weather, is an
example. The assessment of the impact of publicity
(kind and scale) in case of promotions and product
introductions is another example. Secondly, expected
future demand is never a single value, but rather a
spread. If you were to discuss expected demand for a
certain product with a group of experts, they would
say that demand is expected somewhere in the range
between X and Y.
4.2 Information for Setting Stock
Levels
Given the outcome of the process ‘determining
expected demand’, the job of the stock planner is to
set the target levels such that stock will be sufficient
when demand is at the maximum level of the range,
and stock will not be wasted when demand is at the
minimum level. Depending on the volatility of
demand and on the allowed storage life of products in
stock, the planner can have an easy job (steady
demand, longer internal storage life) or an impossible
job (highly volatile demand, very short storage life).
Impossible situations are not primarily the problem of
the planner however, especially if this is a recurring
issue. It is a problem of setting realistic norms for
planning, probably for someone higher up in the
organisational hierarchy.
It gets interesting in the border area when the
Sixth International Symposium on Business Modeling and Software Design
34
planner has a difficult but doable job. In this area the
planner must be creative and use all his available
knowledge and information sources. The planner
might collect further information from experienced
demand experts in order to reduce the expectation
spread, or organise extra production capacity in order
to react quickly (quicker than normal production
schedules would allow) in case of impending stock
shortages. The planner will juggle with delivery risks
and production reaction times in order to find his
solution. Background information about what was
possible or impossible in comparable situations in the
past and informal discussions play a major role in this
kind of decisions.
4.3 Information for Setting Production
Levels
The third process must set production levels such that
production efficiency is optimised. Lot size and
production capacity are the primary determining
factors. Both factors represent discontinuity. Often,
products (or semi-finished products) are optimally
produced in fixed amounts, due to the capacity of
machinery. Optimal operation times for production
lines vary in units of x hours (8 hours is a typical
value), due to work schedules (people work in shifts).
The combination of lot size and optimal operation
times will lead to a production mix that satisfies the
minimum and maximum stock levels set in the
preceding process. In standard situations this can all
be calculated according to fixed patterns and decision
rules. In some situations not all requirements can be
met at the same time. In these cases additional
information is needed, either in finding ways to
squeeze a little more out of production, or in making
sure that some products will be produced in the
required quantity, or in assessing the weight of the
different norms in the given situation and accepting
the additional risk or additional cost.
4.4 Sign Systems
Considered from the viewpoint of the process logic as
analysed above, the information about expected
demand, stock level planning and production-level
planning seems pretty straightforward and a perfect
fit for the domain of formal sign systems. Essentially
it is about three consecutive datasets with the same
structure: date, timeslot, item code, quantity. In
demand this dataset represents an expectation, in
stock-level planning and in production-level planning
the dataset represents a target.
Considered from the more detailed analysis of the
information actually used in real planning situations,
it will be clear that other kinds of information and
other kinds of sign systems are involved. The
inaccuracies of demand expectation, actual stocks and
actual production output must be dealt with; the
planner works with patterns based on experience and
history; risks and possibilities are discussed between
planning, sales and production; and accompanying
instructions are given to the operators in the shop-
floor processes. Information takes the form of
background knowledge, consulting colleagues, oral
communication, and written notes.
4.5 Idiosyncrasies Revisited
Companies differ from each other in the way they
execute their key processes, and for the production of
fresh food for retailers the finished product planning
is such a key process. In this planning process success
on the market (delivery reliability, lead times) and
internal success (minimising production costs) come
together. The way formal sign systems and informal
sign systems are used to support this difficult and
critical process is highly characteristic for each
individual company. Depending on the distribution of
knowledge and experience in one company the ‘real’
decisions and adjustments might be made by
production management (using demand and stock
information), and the planner is no more than a rather
passive Excel-driver and data cruncher. In another
company, the planner plans and the shop floor
executes. In a third company, production and
management meet each day in order to prevent
upcoming problems and smooth out existing
problems. Each company can either flourish or be
ailing. It all depends on the quality and the fit of the
information to and from the planning process and on
the quality of the persons who make the decisions.
5 SALES & OPERATIONS
PLANNING
The subject of the case is the coordination between
production and expected demand using finished
product planning, an area termed Sales & Operations
Planning in the literature and for which ERP systems
provide support. It is then useful to look at what the
literature says about this and to what extent literature
can support the analysis of the case. First a definition
from internet: “Sales and Operations Planning
(S&OP) is an iterative business management process
that determines the optimum level of manufacturing
Hard Problems and Soft Information
35
output”. This definition fits the theme of this paper,
although the term ‘optimum’ in the definition is an
empty shell without criteria and it is a planning
process rather than a management process. The
definition proceeds to state that “The process is built
upon stakeholder agreement and an approved
consensus plan. To help stakeholders agree on a plan
of action based on real-time data, S&OP software
products include dashboards that display data related
to equipment, labour, facilities, material and finance.
The purpose of the dashboards is to provide the
stakeholders with a single, shared view of the data”.
This is not true in the situation considered here. The
planner has a delegated responsibility to solve the
planning problem within the set norms and to signal
when he is structurally unable to meet the norms.
Occasional deviations are permitted (and delivery
reliability prevails over costs), structurally both
norms need to be met. Determining tight but
achievable norms is a mutual undertaking in which all
stakeholders are involved and in which at least
commitment, if not consensus, needs to be achieved.
Operational planning is very different in nature. This
holds even more strongly if Sales & Operations
Planning is not done on a monthly basis, but, as in the
case, must be done on a weekly and daily basis.
The definition in the APICS dictionary, which
should be an authoritative source given the status of
APICS as an organisation (“the premier professional
organisational for supply chain management”,
according to its website, with over 43000 members
and more than 300 international partners), provides
even less of a guide. Because of the language used I
will cite the very long lemma in full:
(APICS Dictionary, 2008, p.121f)
“Sales and operations planning – a process to
develop tactical plans that provide management
the ability to strategically direct its business to
achieve competitive advantage on a continuous
basis by integrating customer-focussed marketing
plans for new and existing products with the
management of the supply chain. The plan brings
together all the plans for the business (sales,
marketing, development, manufacturing,
sourcing, and financial) into one integrated set of
plans. It is performed at least once a month and is
reviewed by management at an aggregate (product
family) level. The process must reconcile all
supply, demand, and new-product plans at both
the detail and aggregate levels and tie to the
business plan. It is the definitive statement of the
company’s plans for the near to intermediate term,
covering a horizon sufficient to plan the resources
and to support the annual business planning
process. Executed properly, the sales and
operations planning process links the strategic
plans for the business with its execution and
reviews performance measurements for
continuous improvement. See: aggragate
planning, production plan, production planning,
sales plan, tactical planning.”
This is a definition (or description) of everything and
therefore of nothing. Why should Sales & Operations
Planning not be about the common daily operational
practice of coordinating demand and availability and
about no more than that? What does something like
“a process to develop tactical plans that provide
management the ability to strategically direct the
business …” add to our understanding of the
problem?
Donald Sheldon writes in his World Class Sales
and Operations Planning (co-published with APICS)
“The S&OP process can have a major impact on the
management of inventory” (Sheldon, 2006, p. 29). He
then devotes chapters to “Creating the Demand Plan”
and to “Operations Planning for the S&OP Process”.
For Sheldon the S&OP process is the coordination
between the various subplans (“Stated in its simplest
terms, the S&OP process is a monthly planning cycle
where plans for both customer expectations and
internal operations are reviewed for accuracy, process
accountability, lessons learned, and future risk
management”, Sheldon, 2006, p. 2), where it should
be essentially about the planning process itself. Of
course there is an important role for higher level long
term planning in companies to coordinate market
developments, production capacities and resource
needs. In this kind of higher level coordination
operational norms must also be determined and
adjusted, and possible measures should be agreed
upon to ‘land’ changed norms with the relevant
internal and external stakeholders. Donald Sheldon
recognises the subordinate role of software: “All that
is needed is a spreadsheet and good problem-solving
tools and skills” (Sheldon, 2006, p. 15). The question
remains, however, where the information for this
problem solving will come from, and how to organise
the different kinds of information flows (both formal
via systems and informal via humans).
Robert Davis analyses what he calls the push-pull
hybrid for supply chain management. “This hybrid
model is based on the premise that you push produce
and pull distribute” (Davis, 2016, p39). This analysis
matches what was described above as the structure of
the problem. His further analysis concentrates on
what is happening in the supply chain as a whole. The
chapter about inventory optimisation discusses the
development of inventory policies that can be
Sixth International Symposium on Business Modeling and Software Design
36
translated into algorithms and executed
automatically. This approach does address the
problem of how to develop ways of coping with the
problem of inventory levels, but it does not address
the problem of the individual planner who uses
information and who makes decisions.
Shaun Snapp (S&OP in Software) describes the
standard S&OP process as follows: (1) review and
sign off the demand plan; (2) review and sign off the
supply plan, and (3) review and sign off the financial
plan. And he gives the time features as a planning
horizon between 1 and 5 years, a monthly planning
frequency, and a monthly planning bucket. This is not
quite the horizon the companies discussed here are
working with. Planning of inventory levels is not in
his list of plans to sign off, Snapp discusses dynamic
safety stock in the chapter entitled ‘How
Misunderstanding Service Level Undermines
Effective S&OP’. He writes: “Safety stock is often set
in companies by simply allowing individuals to
guesstimate what the safety stock value should be and
then provides them with the rights to make the safety
stock adjustments” (Snapp, 2016, p. 105. This is
followed by the remark that in his experience he never
saw this working well, and “Stock levels should not
be controlled by manually adjusting the safety stock.
Instead, safety stock should be dynamically
calculated and automated, and only changed as a
result of changes in the variability of supply or
demand” (Snapp, 2016, p. 107). The last part of
course is true (from ‘only changed as…’), but the first
part presupposes that variability of demand is
represented perfectly in the computer system, with all
relevant information taken into account. This clearly
cannot always be true. And, if manual adjustment of
demand forecast is allowed, planners very quickly
learn the trick how to adjust demand in order to get
the safety stock level they want.
The ERP systems of course offer solutions for
S&OP. Hamilton in his book about MS Dynamics
AX, paragraph 10.1: “You can automatically
calculate the safety stock requirement based on
variations in historical usage and the desired customer
service level” (Hamilton, 2016, p. 236). A bit further
in the same chapter, in paragraph 10.7: “When using
the min-max coverage code, you specify the item’s
minimum quantity and maximum quantity for each
relevant site/warehouse. The minimum quantity
represents the average daily usage multiplied by the
item’s lead time” (Hamilton, 2016, p. 253). So, you
either have safety stock, taking variation of demand
into account; or you have a min-max policy where
average demand represents the minimum stock
needed? Dickersbach gives in his book “Supply
Chain Management with SAP APO™” the following
structure of the demand planning process (somewhat
shortened in my representation): (1) Forecast; (2)
Check on plausibility of the forecast; (3) Production
planning. Inventory planning is not mentioned at all.
(Dickersbach, 2009)
A picture of confusing, vague and contradicting
terminology in combination with conceptual
weaknesses arises from the works mentioned above.
Neither in the literature nor in the software clear
structures of the problem area are defined. A concrete
example of the translation of this messy approach into
actual customer requirements is the following set of
questions for candidate software suppliers by a
company in the food industry:
“How will stock adjustments automatically
influence production schedule?”
“Sequence of production is determined by the
scheduling process. Disruptions in other
processes (up and down) lead to automatic
rescheduling of production capacity; manual
adjusting to schedule needs to be validated and
recorded as an exception”
“How will the production schedule be adapted in
case of (1) late delivery of raw materials”; (2)
delays in production runs; (3) changes in available
stock caused by quality inspections; (4) rush
orders; (5) break down of production lines;
“Reject of the output of finished product at the end
of the production line, how will the system
adapt?”
These questions show a model-based and reductionist
approach to the sales and operation planning process.
The production system is provided a forecast from the
central ERP system (just one value, no information
about spread), processes this to an optimal production
plan, and any deviation or disruption results in
adjustments to the production plan. A fully
deterministic production is assumed, as well as full
and real time information about the actual situation
on the shop floor.
This approach encounters a number of related
fundamental objections: (1) production is not
deterministic, as a result production results will
always deviate from planning which leads to the next
question: what constitutes a deviation? (2) product
registration never fully coincides with production
reality, the view on the computer screen is in the best
case an abstraction of the shop floor reality (details
are not available or omitted) and in other cases a
distortion of reality (for example by enforcing
classifications in search lists that have not been
sufficiently thought through). And, imagine what this
Hard Problems and Soft Information
37
approach would mean for the primary processes on
the shop floor: a continuous flow of changes in
production planning, which will go at the expense of
effectiveness and efficiency.
6 RECAPITULATION AND
CONCLUSIONS
The problem area is about how to combine agreed
service levels with minimising costs. The Sales &
Operations Planning literature clearly indicates the
comprehensive character of the problem, different
viewpoints have to be taken into account. In
structuring the problem field the literature is less
helpful, firstly because of the confusing use of
terminology, secondly because of the time horizon of
months and years, and thirdly because of the lack of
attention for the day-to-day work and challenges for
the people involved in planning. Software, as could
be expected, does not help either. It offers toolboxes
of statistical instruments, dashboards, and algorithms
without much notion of how to apply which
instrument in which situation. The fact that a planner
is responsible for his decisions, and that the planner
has to combine information from many sources in
order to deal with variance, inaccuracy and conflict of
norms is not discussed in the literature.
Essentially, the problem area is about two main
control loops decoupled by a third intermediate
control loop. The first main loop is about service
levels, available stock and expected demand. The
second main loop is about production output and
efficient production. The intermediate loop is about
stock control, firstly for decoupling variance of
demand from smooth and efficient production
processes, and secondly for dealing with all kinds of
disruptions, deviations and inaccuracies in both the
business processes itself and in available information
about the business processes.
The approach is about finding solutions that do
justice to both the structure of the problem (process
logic, formal sign systems) and to the intricacies of
the particular company that must organise its
information in such a way that its competitive power
and distinctive capabilities are enhanced (how it has
found its own ways of dealing with the challenges,
idiosyncratic characteristics, social sign systems).
Human judgement and human communication must
be combined with computing power.
In terms of the coordination mechanisms of
Mintzberg we see a combination of standardisation
(and formal sign systems) and mutual adjustment
(with social sign systems). Standardisation and rigid
definitions of data help to organise information flows
and to automate the processes of determining
demand, setting stock levels, and setting production
levels in standard situations. Dealing with conflicting
norms in non-standard situations, however, is about
mutual adjustment and human responsibilities.
Squeezing out that little bit extra is about human
creativity and problem solving.
To conclude with Kay: distinctive capabilities are
about the characteristics that distinguishes the
individual company from its competitors, and that can
not be readily copied. The challenge is to find the
right combination of formal and social sign systems
that addresses the needs of actual planners in actual
companies and that builds on the existing competitive
power of the company. The high level talk of the
literature nor the reductive approach of ERP software
helps here much.
REFERENCES
APICS, 2008. APICS Dictionary, APICS, Chicago IL.
Boisot, M.H., 1998. Knowledge Assets, Oxford University
Press, Oxford.
Davis, R.A., 2016. Demand Driven Inventory Optimization
and Replenishment, Wiley, Hoboken NJ.
Dickersbach, J.T., 2009. Supply Chain Management with
SAP APO
TM
, Springer, Berlin.
Grice, P., 1989. Studies in the Ways of Words, Harvard
University Press, Cambridge.
Hamilton, S., 2016. Process manufacturing Using
Microsoft Dynamics AX, Scott Hamilton Press.
Kay, J., 1998. Foundations of Corporate Success, Oxford
University Press, Oxford.
Mintzberg, H., 1979. The Structuring of Organizations,
Prentice Hall, Englewood Cliffs NJ.
Sheldon, D.H., 2006. World Class Sales & Operations
Planning, J.Ross Publishing, Ft. Lauderdale FL
Snapp, S., 2016. Sales & Operations Planning in Software,
SCM Focus Press, Las Vegas NV.
Starreveld, R.W., 1963. Leer van de administratieve
organisatie, Samson, Alphen aan de Rijn.
Suurmond, C.P., 2015. Information Systems and Sign
Systems. In Information and Knowledge Management
in Complex Systems. Springer, Berlin.
Sixth International Symposium on Business Modeling and Software Design
38