MEMOIRE - Jumia
MEMOIRE - Jumia
MEMOIRE - Jumia
END-OF-CYCLE DISSERTATION
Diploma of Ingénieur en Logistique et Transport (ILT)
Topic :
1
DEDICATION
To my family.
2
THANKS
Our thanks go against all the people who participated in the writing of this end-of-study report
and to those whose various contributions have punctuated our course within this illustrious
Institute.
- Dr. Moussa DIABY, Director General of the Institut National Polytechnique Félix
Houphouët-Boigny (INP-HB), for his efforts to provide a conducive learning
environment;
- Dr Moustapha TRAORE, Director of the Higher School of Business and Business
Administration (ESCAE) and the entire Management of the School for the quality of
the training received during the three (03) years;
- Dr Losseyni BAKAYOKO, Former Director of the Higher School of Business and
Business Administration (ESCAE) for his dedication and advice;
- Teachers, the technical and administrative service of the INP-HB, and particularly
those of the ESCAE, for knowledge and training;
- Mr. KADIO Jean-Didier, Director of Studies at ESCAE, who accompanied us
throughout the course;
- Dr. ABBE Michelle, Teacher-Researcher at INP-HB and our supervising professor;
- The faculty and administrative staff of the said Institute for the proximity and support
during all these years;
- Mr. SYLLA Mori, Supply Chain Manager of Jumia CI, for assistance and leadership
lessons;
- Mr. OUATTARA Seydou, Warehouse Manager and Internship Supervisor, for
technical assistance;
- All our various collaborators, without whom the substance of this report would be
irrelevant.
3
FOREWORD
As a major pillar of any nation's development, youth have always been at the heart of the
concerns of our African states. Knowing that it was necessary to invest in the education and
training of its populations to ensure better economic, cultural and social development, our
leaders considered it essential to create infrastructures of excellence in the field of education
and training.
Côte d'Ivoire, not wanting to remain on the sidelines of this trend, has set up several didactic
infrastructures for the training of the Ivorian elite, namely:
For several years, these different academic structures have enabled Côte d'Ivoire to brilliantly
train competent executives. In 1996, the Ivorian government decided to centralise the training
skills mentioned above with a view to creating a centre of excellence. Thus, on September 4,
1996, the Ivorian State decided, by decree 96-678, to merge these grandes écoles to create this
center of excellence called Institut National Polytechnique Houphouët-Boigny (INP-HB) of
Yamoussoukro.
In addition to these schools, there are the Classes Préparatoires aux Grandes Ecoles (CPGE),
the Ecole Doctorale Polytechnique (EDP), the International Data Science Institute (IDSI), the
Regional Centre for Higher Education in Metrology and the Aeronautical Training
Centre. These different grandes écoles are responsible for training students in the fields of
administration, commerce, management, mining, industry, agronomy and civil engineering.
4
As for our school, the Ecole Supérieure de Commerce et d'Administration des Entreprises
(ESCAE), it trains senior technicians (Bac +3/DTS) and design engineers (Bac +5).
- Two (2) SDR tracks: Finance-Accounting (FC) and Commerce and Business
Administration (CAE);
- Four (4) design engineering courses: Higher Accounting Studies (ECS), Higher
School of Commerce and Business (ESCA), Hautes Etudes en Assurances (HEA),
Logistics and Transport Engineers (ILT).
The Logistics and Transport Engineers (ILT) training is the one we have followed. To obtain
the diploma, all students must complete their training at the end of the cycle with a stay in a
company in order to combine the theoretical knowledge acquired with the realities of the
company. Thus, the student is required to complete an internship of at least four (4) months
culminating in a dissertation that will be defended before a jury composed of competent
professors. With this in mind, we have been welcomed for an internship since December 12,
2022 at JUMIA-CI, a company specialized in online sales (e-commerce). We were therefore
asked to reflect on the subject: "Improvement of the order preparation process in an e-commerce
company: the case of Jumia-CI."
5
ACRONYMS AND ABBREVIATIONS
AFNOR: French Agency for Standardization
JIT: Just-in-Time
MS: Microsoft
6
LISTE OF TABLES
Table 1 Statement of comprehensive income ....................................................................................... 16
Table 2 Delivery times vary according to destination .......................................................................... 19
Table 3 Relation Nombre d'agents/Coûts de la main d'oeuvre ............................................................. 22
Table 4 Timelines on working day according to the type of goods ...................................................... 26
Table 5 Breakdown of labour costs over the last two quarters in FCFA .............................................. 44
Table 6 Quantity of packaging used for order picking ......................................................................... 45
Table 7 Percentage of LeadTime over six (06) weeks ......................................................................... 45
Table 8 TIER1 performance over six (06) weeks ................................................................................. 46
Table 9 Pourcentage de commandes consolidées ................................................................................. 47
Table 10 Forecast by Average .............................................................................................................. 50
Table 11 Volume processed over the same period ............................................................................... 50
Table 12 Supply Management Policy ................................................................................................... 57
Table 13 Consommation d'emballages par référence ........................................................................... 58
Table 14 Level of alert stocj (𝑺𝒂) ........................................................................................................ 59
Table 15 Determination of the level of replenishment by reference .................................................... 60
7
LIST OF FIGURES
Figure 1 Organigramme de Jumia Côte d'Ivoire .................................................................................. 17
Figure 2 Classification des catégories de produits en pourcentage ..................................................... 18
Figure 3 Order Processing designed ..................................................................................................... 25
Figure 4 Ishikawa Diagram Model ...................................................................................................... 37
Figure 5 Mapping the Causes of Poor Packaging Management Using the Ishikawa Diagram ............ 52
Figure 6 Daily flow of item from 01/01 to 01/04 ................................................................................. 53
Figure 7 Forecasting model on Excel ................................................................................................... 55
8
GLOSSARY
Assortment: All the products and references offered for sale in a store or in a store department.
Packaging: All the material elements which, without being inseparable from the product itself,
are sold with it in order to enable or facilitate its protection, transport, storage, shelf
presentation, identification and use by consumers.
Last mile: logistics operations set up between the warehouse and the final recipient.
Picked: The status of an item picked using a PDA from the storage bins within the warehouse.
Relay point: place of temporary deposit of goods awaiting collection by the final consignees.
Unit load: basic storage and transport unit arranged on a stand for optimal handling
9
BRIEF
More than a decade after its founding, JUMIA seems to be very far from breaking even. The
causes of this underperformance are based on a lack of knowledge of the specificities of the
markets it serves, a difficult control of logistics networks and previous colossal marketing
expenses, the latter of which did not generate the expected level of demand. Based on this
observation, it was a question of rationalizing operating expenses, including those relating to
order preparation. Reducing operating losses means reducing the costs generated by logistics
operations by outsourcing certain activities deemed budget-intensive and refocusing on the core
business, better marketing targeting, and controlling logistics processes within the distribution
center, in particular by improving the order preparation process – an objective to which we are
committed. To do this, a qualitative and quantitative approach allowed us to detect the various
bottlenecks undermining the fluidity of this process. At the end of a discussion group and
through the use of the Ishikawa diagram, the use of the QQOQCP method and an aborted use
of the DMAIC method, we were able to highlight the causes and problems affecting the
operations of the Jumia distribution center, locate the responsibilities, measure the impacts that
the non-resolution of these could generate. We have come up with analysis results according to
which an improvement in the order preparation process requires better predictability of the daily
flow to the company, the implementation of a packaging stock management policy and better
team management. The proposals made are based on order forecasting using the simple
exponential smoothing method, a policy for managing packaging stocks by reorder point and
level of replenishment, and making team leaders accountable.
10
RÉSUMÉ
Plus d’une décennie après sa création, JUMIA semble être très loin du seuil de rentabilité. Les
causes de cette sous-performance s’articulent autour d’une méconnaissance des spécificités des
marchés que l’entreprise dessert, une difficile maîtrise des réseaux logistiques et de précédentes
dépenses colossales en marketing, ces derniers n’ayant a posteriori pas suscité le niveau de
demande espéré. Il s’est agi, à partir de ce constat, de rationaliser les dépenses d’exploitation
dont ceux relatifs à la préparation de commandes. Amenuiser les pertes d’exploitation revient
à réduire les coûts que génèrent les opérations logistiques par une externalisation de certaines
activités jugées budgétivores et un recentrage sur le cœur de métier, un meilleur ciblage
marketing, une maîtrise des processus logistiques au sein du centre de distribution notamment
par l’amélioration du processus de préparation de commandes – objectif auquel nous sommes
dévolus. Pour ce faire, une approche qualitative et quantitative nous a permis de déceler les
différents goulots d’étranglement sapant la fluidité de ce processus. A l’issue d’un groupe de
discussion et par l’utilisation du diagramme d’Ishikawa, l’utilisation de la méthode QQOQCP
et par une utilisation avortée de la méthode DMAIC, nous avons pu faire ressortir les causes et
problèmes affectant les opérations du centre de distribution de Jumia, situer les responsabilités,
mesurer les impacts que la non-résolution de ces derniers pourraient engendrer. Nous avons
abouti à des résultats d’analyse selon lesquels une amélioration du processus de préparation de
commandes passait par une meilleure prédictibilité du flux journaliers s’adressant à l’entreprise,
une mise en place d’une politique de gestion des stocks d’emballages et d’un meilleur
management des équipes. Les propositions faites s’élaborent autour d’une prévision des
commandes par la méthode du lissage exponentiel simple, une politique de gestion des stocks
d’emballages par point de commande et niveau de recomplètement et une responsabilisation
des chefs d’équipe.
11
GENERAL INTRODUCTION
"The only certainty is that nothing is certain."[1] Few jokes best reflect JUBIA's financial
situation ten (10) years after launching in Côte d'Ivoire. Since its creation, the company has
almost never been able to grasp the key success factors of this new field in the still embryonic
economic and technological environment that Africa represents. Between the lack of street
addressing, the road nightmares in the provinces, the problems related to online payment and
the difficult connection to establish with the offline and illiterate fringe of the population,
JUMIA had to refocus its strategic plan around a few clear objectives: restructuring of the
company's management board by first dismissing its founders, Better marketing targeting,
improved assortment and excellent and efficient management of logistics operations.
These logistics operations, which have been inefficient for a long time, have been a reflection
of the group's global strategy of intensive and massive penetration in Africa, thus neglecting
the specific characteristics of each region and country. In order to remedy these concerns, the
general management has undertaken a disciplinary approach in terms of logistics costs through
a plan to optimize the management of the distribution center. There was every reason to believe
that the challenge of e-commerce was not limited to conquering the last mile. However, the
management of the distribution center, in case of under-optimization, results in increased labor
costs, delivery delays, processing errors, and any other deviations in procedures that can lead
to unmet demands.
It is therefore a question for any retail company to control the organization of all the flows of
materials and information evolving within its borders in order to streamline the processing of
orders, optimize the location of references and facilitate the shipment of goods. With
knowledge of the facts, it is legitimate to ask: "How to improve the order preparation process
in an e-commerce company: the case of JUMIA Côte d'Ivoire?"
In order to carry out this study, we set out to answer the following questions:
The aim here is to allow for better order preparation so that periods of high flows are as
controlled as those of reduced intensity. The general objective is to improve the performance
of order preparation operations according to the established performance indicators. In doing
so, we will present the existing tools and methods in order to carry out this study, identify the
12
problems related to the order preparation process and make a proposal for solutions at the end
of the study.
For a scientific and objective approach, scientific analysis tools such as exponential smoothing
forecasting, QQOQCP and Ishikawa diagram analysis methods were successively used to
address these different issues.
Our study is composed of two (02) parts: the first consists of presenting the framework of the
study will serve to present the company and to lay out a theoretical conception of the
phenomenon. The second, on the other hand, concerns the empirical study of the subject. The
latter will present the status quo and its impact on the customer experience, and then put forward
our proposals for solutions.
13
PART ONE : GENERAL FRAMEWORK OF STUDY
This part has two chapters: the first introduces the company, its markets, vision and mission,
and provides a theoretical approach to the concept of the study. It then sets the theoretical scene
by highlighting the methodological framework adopted, preceded by a definitional approach to
key concepts.
14
CHAPTER I: GENERAL FRAMEWORK OF THE STUDY
This chapter is divided into two (02) sections: the first presents the company and its various
strategic business divisions, while the second describes the environment and the host
department.
Launched in 2012, the company claims over eight (08) million active consumers on the African
continent with nearly one hundred thousand (100,000) sellers enjoying this market, over 50
million products and a presence in ten (10) African countries (Ivory Coast, Senegal, Ghana,
Nigeria, Morocco, Algeria, Egypt, Tunisia, Kenya, Uganda) . Despite these figures, the group
is struggling to break even almost twelve (12) years after its creation.
Listed on the New York Stock Exchange (NYSE) on April 12, 2019 with a share price of around
fifteen US dollars ($15), the stock was worth just $2.98 on the evening of 28/01/2024 - a drop
of almost 80% in its initial value. Faced with this decline and the urgent need to turn the
company's accounts around, Jumia's supervisory board has appointed Francis Dufay, previously
in charge of the Ivory Coast subsidiary, as head of Jumia -Group.
15
Table 1 Statement of comprehensive income
The tone is set: the aim is to better manage the costs of the various strategic business divisions
and lay the foundations for profitability. Let's take a look at these divisions.
16
service providers. All logistics partners and assets are seamlessly integrated and managed
through our proprietary technology, data and processes.
Its strategic business divisions are aligned with those of the group. These are Jumia
Marketplace, Jumia On-Demand Service and JumiaPay.
Organigramme
Figure 1 :Organigramme de de Jumia
Jumia CôteCôte d’Ivoire
d'Ivoire
17
consumers. The following table provides a classification of the categories present on the
marketplace by volume
Of the products present on the marketplace and visible on the site (www.jumia.ci), only those
stocked in the Jumia warehouse benefit from the express service. All orders for products
belonging to third-party sellers who manage their own stock are processed within a different
timeframe, depending on the destination city.
18
Table 2 Delivery times vary according to destination
1.2.4 JumiaPay
JumiaPay is part of the Group Executive Committee's overall strategy to offer a fast, secure
payment solution, initially for purchases made on the marketplace. This mission was then
transformed to incorporate all digital payment services, in line with the public's expectations.
These services extend to online purchases on shopping sites other than Jumia, the purchase of
prepaid iTunes, Spotify, etc. cards, and gift cards for games, all in local currency (FCFA). In
19
addition, the development of a technological finance product enables the company to be much
more efficient in settling transactions by reducing the time and resources previously required,
to improve the customer experience by reducing human and therefore emotional intermediation
in online payment processes and by the presumed instantaneity of refunds, to enter new markets
and finally to increase revenue.
- Inbound Logistics : Jumia's inbound logistics include sourcing and purchasing goods,
receiving them on time and in their entirety, unpacking them, checking their quality,
and creating stock keeping units (SKUs) for each vendor and product batch to facilitate
their future location, identification and traceability;
- Storage and stock maangement : Jumia's managers are responsible for storing the goods
received in valid storage conditions, entering all the details into the integrated order
management tool (Order Management System for short, OMS), conducting continuous
inventories and ensuring FIFO (First In, First Out) processing of products or FEFO
(First Expired, First Out) for perishable products;
- L’Order Processing : Order processing refers to the actual preparation of orders. This
processing is mainly limited to Jumia Express items and those that are managed on a
just-in-time basis. This last shipping model, called cross-docking, is a way of organizing
logistics flows that consists of leaving it to suppliers to manage its stock, shipping to
the processing area so that the company can proceed with the allocation of these
20
products in just a few hours. The various operations carried out within Order Processing
are the extraction of items from their location bins, the consolidation of those belonging
to the same order, the packaging and packaging of these items.commandes par un choix
approprié du type d’emballage, un contrôle qualité et conformité final et un étiquetage
pour une meilleure traçabilité.
- La logistique de retour : Jumia's after-sales logistics department handles orders and
items that have been subject to a failed delivery due to cancellation by the customer or
for damage and damage subsequently found on the shipments concerned, etc. Analyses
are carried out after receipt at the department in order to determine the level of damage
in the case of customer returns, to ensure the good quality of the items in the case of
cancelled orders, to put them back in stock for online or offline sale depending on
whether these items are more or less damaged and to proceed with their destruction in
the case of unsaleable products.
21
- A better orientation of activities leading to the allocation of resources and capital to key
areas with interesting and obvious returns on investment for the entire economic
ecosystem. This measure is coupled with the closure of certain strategic business areas
such as "Dark Stores" in April 2022, on-demand service that offered basic necessities
to consumers with a maximum delivery time of fifteen (15) minutes ;
- A fundamental approach to growth that, historically, has been drained by colossal
spending on marketing. Instead, the new council will improve the offer and price of
goods and services on the platform, while adopting better marketing targeting;
- Finally, a high level of cost discipline in order to achieve efficiencies across the entire
structure, particularly in the areas of general and administrative expenses, distribution
center costs and labor. As part of these streamlining efforts, the group significantly
reduced its presence in Dubai, where some management functions were located,
reducing the workforce by more than 60%. Most of the remaining employees have been
transferred to the African offices, closer to consumers, sellers and operations. We
anticipate that these workforce reductions will allow us to generate personnel cost
savings in 2023 compared to 2022. The Group's financial statements will be presented
in Appendix 1.
2.2.2 Mission
Jumia Côte d'Ivoire's situation aligns perfectly with that of the group. The financial discipline
to which the various subsidiaries of the group must be bound inexorably involves better
management of the workforce and the costs it generates, rigorous management of packaging
and better monitoring of the performance indicators adopted for efficient order preparation.
Needless to say, Jumia's distribution center is the company's largest cost center, with the
fulfillment team accounting for approximately 30% in value and volume. Enabling better
control of these costs is the purpose of our study.
22
CHAPTER II : THEORETICAL AND METHODOLOGICAL
CONCEPTION OF THE PHENOMENON
This chapter is dedicated to the definition of the different concepts addressed during this study
as well as to a global and scientific view of the subject. In it, we clarify the terms related to the
host departments, in this case the Supply Chain department, as well as the different traditional
conceptions of logistics, supply chain, e-commerce and their main reciprocal interactions.
- The collecting or picking of single items : This is the operation by which the items for
which an order order has been collected are collected. There are different picking
methods, chosen on a case-by-case basis depending on the situation the company is
facing. The first method is related to cross-docking: once the items that are the subject
of an order arrive in the warehouse, they are processed quickly due to the short transit
time to which they are subjected, in particular by their identification, labelling and
dispatching according to their delivery locations. For Jumia Express items, subject to
"warehousing" » At Jumia, the main picking operation chosen is "Man-to-goods"
23
because of the warehouses that are still very humanized and less robotized. This manual
operation carried out by a picker agent consists of this agent going through the storage
area in order to collect the different products belonging to an order order. The latter,
under the order of his team leader, can collect by dedicated location area, by reference
(Stock Keeping Unit) in order to proceed with a subsequent distribution of these items
by orders or the picker agent can proceed to a collection by orders. Depending on the
order flow, different pickers are assigned to different methods in order to speed up the
order processing time ;
- Consolidation of orders: this operation consists of grouping together the units of the
references previously picked in order to form a complete unit load to be shipped; This
is in order to form a single consolidated shipment of the items belonging to the same
order and thus eliminate the logistical costs that would be generated by an incomplete
or multiple delivery of the same order. This operation, if not controlled, could lead to
customer dissatisfaction and therefore exponential logistics costs. The rise of e-
commerce has been coupled with a 33% increase in the rate of returns. It is therefore
essential to group the items belonging to the same order so as not to incur additional
logistics costs. Some valid reasons for not consolidating certain orders are: items of
different sizes belonging to the same order (an order consisting of a pen and a
television), items from categories that are difficult to consolidate ;
- Packing: This is the last operation that makes up the processing of orders and therefore
the one responsible for making up for potential processing errors made during the first
two operations. This operation consists successively of carrying out a quality and
conformity check of the various items included in the order being processed, of choosing
the right packaging in accordance with the packaging policy adopted which has been
determined according to the nature of the products to be packaged, their value, their
degree of resistance, the means of transport used for their shipments, the various
successive manipulations to which the products are subjected. In addition, this order
processing department is also in charge of labeling orders once packed in order to
facilitate their future traceability and the modification of their status depending on
whether they are being delivered, delivered, in transit or in return transit. Although it is
an expensive investment, the importance of packaging can therefore be measured by the
advertising relay it represents, the distinction it generates vis-à-vis other market players
by its colours and ergonomics and its imperative importance in securing goods when
24
they are shipped.
Source : Mecalux.
- The lead time : It corresponds to the time that elapses between the order of an order and
its availability to the sorting department without a break in load. This time includes the
times of the various order processing processes: collection or picking, consolidation of
25
items belonging to the same order, their final packaging. This time shall not exceed four
(04) hours of time;
- Cut-off: The promise that Jumia is committed to keeping to its customers is that they
will have orders placed on Jumia Express (JE) goods in record time. These timelines are
set out in the table below :
JUMIA
NON-JUMIA EXPRESS
EXPRESS
TIER 1 : Abidjan-Lagunes - ABOBO - ADJAMÉ - COCODY -
YOPOUGON - BINGERVILLE - MARCORY - KOUMASSI - 1 2-3
PLATEAU - PORT-BOUET
TIER 2 : Aboisso - Agboville - Anyama - Azaguié - Bouaké -
Dabou - Grand-Bassam - N'Douci - Sikensi - Tiassalé – 1-2 3-4
Yamoussoukro
TIER 3 : Abengourou - Adzope - Agnibilekrou - Agou - Akoupé
- Bangolo - Bondoukou - Bongouanou - Bonon - Bouafle -
Daloa - Daoukro - Dimbokro - Divo - Duekoue -
Ferkessedougou - Gabiadji - Gagnoa - Guiberoua - Issia -
Katiola - Korhogo - Kotobi - Koun Fao - Lakota - Man - Meagui
2-3 3-4
- Niakara - Oupoyo - San-Pedro - Sassandra - Seguela - Soubre
- Tafire - Tanda - Tiebissou - Toumodi - Vavoua - Yabayo –
Fresco
TIER 4 : Adiaké - Arrah - Béoumi - Bettie - Biankouman -
Blolequin - Bocanda - Bouna - Boundiali - Buyo - Danané -
Diawala - Didievi - Diégonefla - Djekanou - GrandBereby -
Grand-Lahou - Grand-Zattry - Guiglo - Jacqueville - Kani - Kong
- Kouto - Mafere - Mankono - M'Bahiakro - M'batto - 4-6 5-7
M'Bengué - Nielle - Odienne - Ouangolo - Ouellé - Oumé -
Ouragahio - Saioua - Sakassou - Sinematiali - Sinfra - Domaine
SOGB - Tabou - Tai - Tingrela - Touba - Toulepleu - Transua -
Yakasse Attobrou - ZOUAN-HOUIEN – Zuenoula
It is therefore important in order to be on the next-day delivery to control our order preparation
times and to be in tune with the purpose of the logistics, to make the goods available at the right
time, in the right place, with the right quality and at the right cost for both the customer and
Jumia.
- Consolidation: This indicator measures the rate of consolidated orders. An order is said
to be consolidated when all the items in the order are packed under the same package.
26
This is used to deliver orders in a single delivery, to reduce packaging costs per order,
to guarantee customer satisfaction;
- The productivity of the various agents carrying out the operations (picking and packing)
during the different working periods is also measured. It is used to know on a daily basis
and over a semester the optimal number of people to recruit in order to perfect the
processing of orders. The work of these agents is governed by Ivorian labour law with
fixed-term contracts with unspecified terms.
All these operations are optimized by an order management software package called OMS. It
makes it possible to manage stock management and order tracking while having a real-time
view of the various flows. The various dashboards used to monitor the variation of these
indicators are presented in Annex 2.
1.2.1 Logistics
Attributing the origin of logistics to the military domain is a platitude that never ceases to
resound in lecture halls. However, few have attempted to enumerate the legacy of this scientific
corpus to the various industries that cannot do without it, as well as to display the historical
development it has had mainly during the twentieth century.
According to Yves Pimor and Michel Fender, logistics according to its military origin
represents "the practical art of moving armies and supplying them by establishing and
organizing their supply lines." Thus, the mobility of troops and their supplies are the basis for
the effectiveness of operations. This necessary agility was reinforced during the Second World
War (1939 – 1945) by the possibility of supplying the troops with food and weapons through
the exclusive deployment of real logistics chains made possible by the development of road
transport, particularly by rail, sea and air transport. The outbreak during this period of
operations research facilitated the resolution of some of the bottlenecks faced by the various
operations managers. These included inventory management, optimization of the sizes of the
convoys, delivery rounds, construction of temporary infrastructure (warehouses, distribution
logistics platforms, unloading bridges, etc.)
At the end of this war, the prosperous period of growth and industrial development that
constituted the thirty glorious years was marked by an organization of work in perpetual search
27
for productivity. The parcellarization of tasks in the Smithian sense of the term has been the
subject of a major and all-out adoption in the course of the production operations of enterprises.
The aim was to produce without waste, loss and dead stock by breaking down tasks into
elementary and timed operations. This scientific organization of work (OST) led to a
standardization of finished products which, later, contributed to the end of this cycle of great
prosperity; The first oil crisis of 1973 contributed to a slowdown in economic growth and thus
to a decline in productivity gains. This purge of capitalism has led to one of the most beautiful
revolutions in the logistics field: the "just-in-time" of the Toyota School.
Faced with the contraction of demand and an increase in the number of finished products in the
automotive industry, the major objective of automotive manufacturers is becoming to produce
when consumers demand it and how they demand it; To do this, there are two proposed
solutions: the first is to improve demand forecasts through a viable statistical model over a short
time horizon and the second is to trigger a replenishment following an order order, a pull
production with minimal stock. This triggering of post-upstream production by an expression
of the need by the post-upstream allows a reduction in the cost of production by keeping a
strictly minimal stock used to compensate for any malfunction within the production process.
The generic "just-in-time" model makes it easy to apply to a wide variety of areas of the supply
chain and related service activities.
This chronological development of logistics has gradually put the customer experience at the
center of its key success factors. By moving from strictly productive and profitable industrial
production to thrifty customer orientation, logistics has asserted its intrinsic relationship with
marketing; The latter offers distribution a prominent place in its means of action.
In short, we will retain the definition of logistics proposed by John J. Coyle, Edward J. Bardi
and C. John Langley: « Logistics is to ensure the availability of the right product, in the right
quantity, and in the right condition, at the right place, at the right rime, for the right customer,
at the right cost. »1
1
John J. Coyle, Edward J. Bardi, C. John Langley, The Management of Business Logistics, West Publishing
Company; 5th ed., 1992.
28
1.2.2 Supply Chain
The customer orientation at one with logistics has made it possible to overcome the prolixity of
the academic literature in describing the "Supply Chain" phenomenon. We will make sure to
limit it according to the needs of our practice.
This anglicism, whose literal translation of supply chain, does not allow us to grasp its full
extent, breaks with the division of labour by calling into question most of its foundations. The
supply chain, which post-Just-in-Time, appeared at the end of the 70s and really became more
popular in the early 2000s thanks to Martha C. Cooper, Douglas M. Lambert and Janus D.
Pagh2.
More than a new end for logistics, the supply chain heralds a break with the logistics dynamics
of yesteryear, which was built on a blind quest for productivity and therefore a thorough
parcellarization of tasks. To this end, the supply chain offers integrated process management,
coordination of a set of activities over time in order to produce value to the external customer
measured in time, quality and cost. The condition of the efficiency of a company through the
scientific organization of work gives way to a need for efficiency due to a much more
competitive environment and consumers less indifferent to standard products.
The main objectives of the company's main objectives was to increase the speed of movement
of products within the chain, to obtain a competitive advantage3 through the interoperability of
these different processes and to resolve any distortions in demand analysis as we move away
from the end user4.
If Smith's division of labor is called into question, willy-nilly, we could bring out the entire
wide range of activities to which supply chain managers are devolved – in e-commerce in
particular. It encompasses all the steps necessary for products ordered online by customers to
be manufactured, stored, packaged and delivered to their final destination.
To this end, the supply chain includes the planning and management of all activities related to
supplier sourcing, procurement and processing, as well as all the logistics activities mentioned
above. This includes, but is not limited to, coordination and collaboration between partners in
2
Cooper, M.C., Lambert, D.M. and Pagh, J.D. (1997), "Supply Chain Management: More Than a New Name for
Logistics", The International Journal of Logistics Management, Vol. 8 No. 1, pp. 1-14.
https://doi.org/10.1108/09574099710805556
3
Michael E. Porter, Competitive Advantage: Creating and Sustaining Superior Performance, Simon and
Schuster, 2008
4
L’effet de Forrester
29
the global chain, which may be suppliers, intermediaries, service providers and customers.
Fundamentally, therefore, the Supply Chain integrates supply management and demand
management within the company and between companies.
Also, managing customer returns is no less important. One would be tempted to object that this
return logistics process is the result of the deployment of e-commerce. In France, 24% of
products sold in e-commerce are returned. The reasons behind these delivery failures are mostly
due to processing errors (invoice reversal, spoiled products, wrong product delivered, lack of
conformity on size, color) or content errors (difference between product sheet and delivered
product).
Finally, it will be assumed that the relevance of a Supply Chain is measured by the degree of
integration of these different components. From now on, in addition to physical flows, it is
necessary to take into account the intra-company and inter-company information flows and the
financial flows of all the different operations that they group together under its name.
It was preceded by historical distribution and mass distribution. In the period before the first
industrial revolution, this first variant was characterized by a sedentarization of trade in
increasingly populated cities or, on the contrary, by the activity of traveling merchants and
itinerant vendors going from town to town to temporary fairs and markets to present their
products. The emergence of these convenience stores has led to a strong degree of
specialisation; Each of them offers a strict family of products (draper, baker, goldsmith, etc.).
Those bodies had exclusivity in the marketing of their products, an exclusivity in return for
which they endeavoured to confine their activities only to the distribution of a limited range of
those categories of products. The distribution channels were longer or shorter depending on the
different product categories served.
30
Direct Channel Court Channel Long Channel
WHOLESALER
DISTRIBUTOR
DISTRIBUTOR
Source : Mercator
According to a report published by GIZ, the distribution sector in Côte d'Ivoire remains
dominated by this form of distribution (more than 85%); There were already more than 16,000
convenience stores across Côte d'Ivoire in 2017. Also, our major markets have remained places
of exchange for different goods and services that still escape the opportunities offered by the
Internet. These include poultry, sea animals, livestock, alcoholic beverages, soft drinks,
children's items, etc.
Since the 1960s, "historical distribution" has increasingly given way to what we now call "mass
distribution". With a strong food dominance, large-scale distribution in Côte d'Ivoire is
characterized by an oligopoly with competitive fringes. According to the National Institute of
Statistics, the Société de Distribution de Tous Biens en Côte d'Ivoire (SDTM-CI), the Ivorian
Company for the Promotion of Supermarkets (PROSUMA), the Compagnie de Distribution de
Côte d'Ivoire (CDCI), the main players in this sector, have been shaken by the recent and
gradual arrivals of the French behemoths Carrefour, Auchan and Casino. Since December 2015,
when the first shopping center "PlaYce" was opened in Abidjan Marcory, housing a Carrefour
hypermarket, the strategy has been to multiply the number of distribution facilities within the
economic capital, first and finally throughout the national territory. This strategy of intensive
penetration is necessitated by the restricted catchment areas served by each distribution format.
31
Table 5 : Average square footage by distribution format
It is easy to see that the size of the assortment offered in a distribution format depends on the
size of the format. This fact is rehashed by e-commerce, especially since, unlike a physical
store, the Internet does not allow any limit on the width and length of a retailer's assortment. In
2024, the world's largest retailer had more than 350 million items listed on its site.
In doing so, in order to overcome the limitations of "brick-and-mortar" shops and to seize all
the opportunities offered by the Internet, e-commerce companies have multiplied over the last
decade on the Ivorian territory. This excitement for this new form of consumption has been
made possible by the emergence of a middle class backed by a mostly young population. These
Digital Natives, whose experience is marked by their addiction to smartphones and social
networks, have forced retailers to rethink their business models and therefore their customer
promises. This was followed by the birth of new kind of businesses :
32
Table 6 : Four (04) categories of e-commerce
Business Consumer
Several players, including Jumia subsidiaries, have failed in recent years, either because of their
business model or because of logistical problems. In December 2016, CDiscount put an end to
its African adventure. Cdiscount's positioning, which is a little more upscale than Jumia, may
not have been able to attract enough customers. Africashop, a subsidiary of Carrefour, has also
closed shop, and more recently Afrimarket and Yaatoo.
In Côte d'Ivoire, several players, including Jumia subsidiaries, have failed in recent years, either
because of their business model or because of logistical problems. In December 2016,
CDiscount put an end to its African adventure. Cdiscount's positioning, which is a little more
upscale than Jumia, may not have been able to attract enough customers. Africashop, a
subsidiary of Carrefour, has also closed shop, as have Afrimarket and Yaatoo. Faced with the
difficulty of making this economic model viable and exogenous factors such as Internet and
mobile penetration, cash on delivery, access to an illiterate segment of the population, the
primary objective remains to rationalize the company's activities and processes.
In addition, it was clearly a question of abandoning activities with poor growth prospects,
drastic reduction of our marketing expenses in the belief that the right investment remains a
perfect customer experience made possible by the excellence of our operations; for the
fulfillment department, better order forecasting and an adapted packaging management policy.
33
SECTION 2 : Study methodology
The purpose of this report is to present the data processing tools and analytical methods used
throughout this report. They are therefore tools for collecting and processing data.
34
- Grafana: Grafana is free software licensed under the GNU Affero General Public
License Version 32 (formerly licensed under Apache 2.0 before April 2021) that allows
data visualization. It makes it possible to know the number of items to be processed,
processed, when and by whom at all times for all the subsidiaries of the group. (See
Appendix 5)
- Qualitative information on specific issues or topics such as the launch of a new product
or the observation of existing processes, or to find out about the concerns of the
community regarding the implementation of a new project;
- Information on what motivates the behaviour of certain actors (consumers, employees,
etc.);
- Information to identify participants' needs ;
One of the advantages of focus groups is the "snowball effect" they cause. Thus, a participant's
answer to a question can elicit responses from other participants, etc. In addition, responses are
more spontaneous and new ideas sometimes spring up through interaction with others. At the
end of the session, once the questions have been asked and the ideas have been exhausted, the
facilitator or researchers will proceed to analyze the comments made and preserved in writing,
on video or audiotape. The following table summarizes a series of aspects that need to be
clarified when setting up such a group.
35
Table 7 : Questions for Organizing a Focus Group
The Cause and Effect Diagram, or Ishikawa Diagram, or the Fishbone Diagram or 5M, is a tool
developed by Kaoru Ishikawa in 1962 and used in quality management. This diagram
graphically depicts the causes leading to an effect.
It can be used as a tool for moderating a brainstorming session and as a tool for synthetic
visualization and communication of the identified causes. It can be used to investigate the cause
of a problem or to identify and manage risks during the implementation of a project. This
diagram is usually structured around the concept of 5M :
- Material : the materials used and involved, and more generally the inputs to the process
;
36
- Measurement/medium: the environment, the positioning, the context.
Source : Etudiant
Each of the causes connected to the five branches and the overall effect must be quantified in
order to allow a hierarchy by a Pareto chart: this will then help to find solutions to eliminate
these causes, according to priorities. The 5 branches are not systematic, they depend on the
problem encountered.Décrire l’effet n’est pas immédiat et facile, il faut rechercher un libellé
précis, concis, chiffré.
37
The approach consists of collecting and analyzing all the relevant information available by
asking as many questions as possible of all kinds regarding a problem, a situation, a subject
defined beforehand. The idea is to think constructively in order to highlight the main causes of
the problem, the reasons for a specific project, define the main lines of an action plan, etc.
- Define :
Objective: Clearly define the problem or opportunity for process improvement.
Key Activities: Identify stakeholders, set measurable goals, define process
boundaries, and clarify client requirements.
- Measure:
Objective: To collect data to understand and assess the current state of the process.
Key Activities: Identify key variables, develop a data collection plan, collect
relevant data, and establish the current performance of the process.
- Analyze :
Objective: Identify the root causes of problems and opportunities for improvement.
Key Activities: Analyze collected data, use statistical and graphical tools, identify
cause-and-effect relationships, and determine sources of variation.
- Improve:
Objective: Implement solutions to solve identified problems and improve the
process.
Key Activities: Generate potential solutions, evaluate those solutions, implement
changes, and monitor the impact of the solutions adopted
- Control :
Objective : Follow-up the precedented initiatives
Key Activities : Develop control plans, implement monitoring procedures, train
staff, and establish performance indicators to ensure the long-term stability of the
improved process.
38
The DMAIC methodology encourages an iterative and continuous approach, where each phase
is interconnected and can influence subsequent steps. It fosters data-driven decision-making
and a methodical approach to solving problems and improving processes.
39
PART ONE CONCLUSION
This first part allowed us to set the scene for our study. First of all, it was a question of
presenting the company that hosts us as well as the department in which our research work was
carried out. Then, the definition of the key concepts made it possible to announce the different
corpora to which our field of research belongs, to review the different challenges of this
operation and to expose the methodology used in order to provide a qualitative and acceptable
response.
40
PART TWO : EMPIRICAL STUDY
The generalities of our study having been set out above, it is now a question of portraying the
current situation of order preparation within Jumia Côte d'Ivoire for Jumia Express and non-
Jumia Express orders, Warehouse or Cross-Docking orders. It will then be a question of making
proposals for solutions in this direction and proposing an implementation plan at the end of the
analysis.
41
CHAPTER III : Presentation of the phenomenon studied
The two sections that make up this chapter will respectively address the current state of the
order picking process and the various limitations that hinder it.
1.1 Le Picking
When an order is placed, customer service contacts the customer to ensure that the customer
has filled in the information according to a well-defined script. Some orders are validated
automatically when the customer concerned is not on his first order and has a 5% lower failure
rate and a success rate of at least 80% on these previous orders.
After validation, the items belonging to the order that is ready for fulfillment are removed from
their bin located in dedicated bin areas. To do this, agents use a PDA (Personal Digital
Assistant), go to the bin location area in which the reference of the item to be picked is located,
beep an article from the part number and then place the item picked in a unit load.
1.2 Consolidation
The consolidation operation only applies to orders with at least two items (Multi order). Orders
with a single order are immediately pushed to the packing lines to be packed and/or packaged
before being handed over to the delivery department. After picking, items should be grouped
according to whether they belong to the same order in order to minimize the cost of packaging
per item and to provide the customer with the object of their order in a single delivery attempt.
With a PDA, the agents in charge of this operation in turn beep each ID they enter into a bin
until all the items belonging to the same order are finally gathered together and ready to be
packed.
1.3 Packing
The purpose of this operation is to protect previously consolidated orders from adequate
packaging. The choice of packaging by the agents in charge of packaging depends on the
quantity of the items in the order, the category to which these items belong and the possibility
of putting them in the same packaging. There are 5 types of packaging:
42
- JUMIA bag with built-in handle (Medium format) 45 cm x 50 cm;
- JUMIA Cut-Out Handle Bag (Small) 35 cm x 30 cm;
- Grey secure envelopes 36.5 Cm x 25.5 Cm + barcode;
- Grey secure envelopes 50 cm x 40 cm + barcode.
The rule behind the packaging is that the package size ratio is S:M:L = 1:5:15. This means that:
The order picking model is substantially identical to the one presented in Chapter 2.
The operations on these different shifts are carried out by workers whose contracts are
assimilated to fixed-term contracts with an imprecise term, a contract held over a period of six
(06) months that cannot be combined. This affects the experience effect that is being wiped out
due to recruits who are always new every six months.
Now let's look at the costs of the different order processing operations.
43
Table 5 Breakdown of labour costs over the last two quarters in FCFA
It appears that :
- The average share of the number of agents of the PO compared to that of Jumia CI is
25.57% in volume and 24.94% in value. In other words, 25% of the relative labour costs
borne by the entire company are due to order picking operations;
- The average share of the number of PO agents compared to that of fulfillment is 34.78%
in volume and 32.50% in value. In other words, one-third of the labor costs incurred by
the entire distribution center are due to order picking operations.
These figures reveal how crucial it is to control the number of agents summoned on a daily
basis in order to best manage the costs they generate.
The quantities of packaging required for packing are not without consequences for our cost
structure. The different types of packaging have consumption that varies according to the order
flow processed over the period considered, the categories of these items. Below is the table on
packaging consumption for the first four (04) months of the fiscal year.
5
FTE : Full-time equivalent means the unit of measurement equivalent to a person - worker or student - a unit
of a workday, applicable in a variety of contexts. In most cases, the FTE formula measures an employee and their
workload.
6
OP signifie Order Processing. OP FTE (Order Processing Full-time Equivalent) refers to the operators required to
carry out the order picking process
44
Table 6 Quantity of packaging used for order picking
Consommation Mensuelle
Description
January February March April
Lot de 25 Sachets Orange imprimé Jumia 64.5 Cm x 52 Cm - Orange 150 155 208 195
Source : L’entreprise
It is difficult to identify a trend through the temporal evolution of the quantities of packaging
consumed. However, there is every reason to believe that their management could be improved.
- Lead Time: Poor daily workforce planning for the various order picking operations
could lead to a variation in the percentage of items processed on time (4 hours). A shift
in reduced staff will hardly be able to achieve the objective assigned to it. On the other
hand, an overstaffed shift will increase costs per order processed.
45
The Green Service Level gives the share of items processed in less than four (04) hours. The
customer promise (cut-off 10 p.m.): this promise assumes that all TIER1 orders placed before
10 p.m. will be delivered the next working day.
To do this, it is a question of processing all the items belonging to these orders and handing
them over to the middle and last mile service before 11:30 p.m. The following is a table
summarizing the performance of this indicator:
There is a lack of control over this performance indicator over the different periods. This will
result in a deterioration of the customer experience, alter the engagement of the lovers of the
platform and the competitive advantage of JUMIA. A period of understaffed work will only
partially help to achieve the goal of processing all orders placed by 10 p.m. before 11:30 p.m.
Also, a period of overstaffing will increase the department's operating costs. The goal is the
same, increase the precision of the staff in order to get as close as possible to Green's 100% on
the customer promise.
7
SLA = Service Level Agreement, A contract between a service provider and a customer that describes the level
of service that the provider will provide. It defines the metrics that will be used to measure the performance of the
service, as well as the corrective actions or penalties that will be imposed if the service does not meet the agreed
standards.
46
Consolidation: The goal here is to pack items belonging to the same order as much as possible;
This is in order to minimize the cost of packaging per item and therefore per order order as well
as to initiate the shipment in a single delivery.
Faced with these almost random performances, it is essential to identify the causes at the root
of this performance gap and then to propose appropriate solutions.
8
Each order after being packed is associated with a tracking number. An order split into multiple packages will
have multiple tracking numbers bijectively.
47
CHAPTER IV : ANALYSIS RESULTS AND PROPOSED SOLUTIONS
This chapter, using the Ishikawa diagram, the Quintilian method, and a strand of the DMAIC
method, responds to the difficulties set out above while concluding with recommendations for
remedying them.
Script Responses
Who to gather All the team Leaders
together ?
How many are they ? 9
Who’s the facilitator ? Myself
How much time ? 3 hours
Guidelines - The aim was to identify all the causes of our
underperformance ;
- Define areas for improvement.
Source : L’Etudiant
At the end of this group interview, it emerged, in accordance with our preliminary study of the
status quo, that the underperformance observed within the department has two main causes:
Poor order forecasting; Packaging management on a case-by-case basis. We analyze these
bottlenecks in turn.
48
1.2.1 Define
In order to clearly identify the characteristics that need improvement, we use the Quintilian
method. The definition of the problem is shown in the following table:
The absence of an order forecasting model leads to variability in performance indicators and a
shortage of consumables, negatively impacting the order preparation process.
1.2.2 Measure
The existing forecasting model is similar to a moving average. It consists of averaging the
number of articles submitted for preparation in the last two weeks of the days concerned
according to the formula below :
In this case, the forecast error corresponds to the difference between the number of picklisted
items and the expected number of items over the corresponding period: ep = Dn - Pn.
49
Forecasting errors close to zero will improve order processing as the daily planning of agents
will also be accurate. The following is a forecast table based on the average:
Last 2 Weeks Average Monday Tuesday Wednesday Thursday Friday Saturday Sunday
Average Productivity /FTE Picking 500 500 500 500 500 500 500
Average Productivity /FTE Packing 700 700 700 700 700 700 700
Source : L’entreprise
Prod. /FTE Picking 441 470 481 493 541 408 438
Prod. /FTE Packing 707 678 682 603 700 599 541
Source : L’entreprise
A large forecasting error (𝑒𝑝 >500) leads to more or less one agent being staffed and therefore
order unpreparedness or additional costs generated.
50
After discussions with the team leaders, the inaccuracy of the forecasts is also due to sales
campaigns or destocking of certain references taken by the salespeople without notice... These
recurring practices do not allow operations to be conducted properly during the various work
periods, as teams find themselves overwhelmed by the flow of orders in these cases.
These causes can be classified and arranged on the Ishikawa diagram as shown in the figure
below:
51
METHOD
PACKAGING INVENTORY
MISMANAGEMENT
STRESSFUL ENVIRONMENT
Unfamiliar with Data analysis tools
Figure 5 Mapping the Causes of Poor Packaging Management Using the Ishikawa Diagram
Source : L’étudiant
52
A consumables management policy is proving to be substantial in order to remedy this. That is
the whole purpose of Section 2.
To do this, let's first represent the daily flows of items submitted for preparation using a curve.
12000
10000
8000
6000
4000
2000
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97
Day Qté
Source : L’étudiant
We can see from this curve that the flow of items varies in a sawtooth pattern during working
days before dropping on Saturdays and Sundays. We have weekly seasonality.
∑100
1 𝑋𝑖
This time series averages m = = 8853 items
100
1
𝑤𝑖𝑡ℎ 𝑎 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑒𝑞𝑢𝑖𝑣𝑎𝑙𝑒𝑛𝑡 𝑡𝑜 𝜇 = √100 ∑100
1 (𝑋𝑖 − 𝑚)² = 1791. According to
the Bienayme-Chebitchev inequality, the probability that datum X deviates from the mean (m)
53
1
de plus or minus k times the standard deviation is less than . Thus, 75% of the values of X
𝑘2
are within two (02) standard deviations of the mean. We will use this ratio as confidence interval
according to which 75% of X ∈[m-2μ ; m+2μ].
The forecasting model used for this time series is exponential smoothing. It calculates an
estimate of the trend and seasonality of the data using a weighted combination of previous
observations.
Exponential smoothing is particularly useful when the data shows trend and/or seasonality, but
it is difficult to determine an accurate pattern to represent it – which is the case here.
Its principle is based on the fact that the most recent observations are generally more important
in predicting future values than older observations. The X data for the month of January has
little consequence on that of the month of April, for example.
We use an Excel spreadsheet to facilitate the forecasting model according to the following script
:
54
14000
12000
10000
8000
6000
4000
2000
0
29
57
85
1
5
9
13
17
21
25
33
37
41
45
49
53
61
65
69
73
77
81
89
93
97
101
105
109
113
117
121
125
Qté Prévision(Qté) Limite de confiance inférieure(Qté) Limite de confiance supérieure(Qté)
Source : L’étudiant
We note that the predicted data are framed by two confidence limits corresponding to the
application of the Bienayme-Chebitchev inequality presented above.
2.1.2 Recommandations
Since simple exponential smoothing has been proposed as a solution to the previous imprecision
of the order flow, it is necessary for the actual flow not to deviate from the confidence interval:
An accurate forecast of the daily flow of orders will result in controlling labor costs due to order
preparation, allowing a better control map of performance indicators, and increasing the
productivity of operating agents through experience.
55
- The stockkeeping system that takes a snapshot of the components of the supply: the
physical stock level, the work in progress or expected (requested but not yet received)
and the due (quantity of packaging already on the packing lines). This is what we will
refer to as the theoretical stock in the rest of the study. Formula : Theoretical stock =
𝑃ℎ𝑦𝑠𝑖𝑐𝑎𝑙 𝑠𝑡𝑜𝑐𝑘 + 𝑂𝑢𝑡𝑠𝑡𝑎𝑛𝑑𝑖𝑛𝑔 𝑎𝑚𝑜𝑢𝑛𝑡 − 𝑆𝑡𝑜𝑐𝑘 𝑑𝑢𝑒 ;
- The inventory management system, which represents all the information and decision
rules used to trigger a demand for packaging in order to best meet the management
objectives adopted by the company.
The stockkeeping system is similar to the information provided by any dashboard, whereas
management does not refer to the steering actions and decision-making resulting from the
information provided by this dashboard.
For our case study, we will limit ourselves to the implementation of decision rules for
consumables.
2.2.1 Solution
Two types of information are necessary for good packaging management: first, the downstream
flow of orders and the constraints related to the company's policy on packaging demand.
The downstream flow of orders has been the subject of our theoretical development in point 2.1
of the same section. We therefore postulate a deterministic demand on the fixed time horizon,
in this case the week.
56
These different modes of operation result in four (04) basic supply management policies which
are presented in the table below.
WHEN ?
T S
We present the advantages and disadvantages of each policy in turn before choosing the most
appropriate one.
- T,Q policy (Replenishment and Quantity Fixed): Its only advantage is the simplicity of
implementation. This non-reactive policy makes the control of the stock and the quality
of service undynamic. At the cost of a fixed quantity requested, in an uncertain
environment, stock control requires an increase in the frequency of replenishment –
which does not allow this policy. It is therefore not appropriate ;
- S,Q policy (Replenishment at the point of order and in fixed quantity): this policy offers
good control of stock-outs, the main bottleneck in our case. In the event of aberrant
flows, the frequency of orders accelerates since the company allows requests at any
time. The calculation of the alert stock can be updated according to the latest demand
forecasts. Its only downside is that it requires a permanent inventory ;
- T,R Policy (Fixed Replenishment by Replenishment Level): The Replenishment Level
helps control overstocking but leads to a higher stock that is less suited to fluctuations
in demand. Knowing that the volume of orders is irregular overall, this policy seems to
be less appropriate in our case;
Policy S,R (Replenishment at Reorder Point and by Replenishment Level): This policy is
similar to the S,Q policy. Suffice it to take for R the sum of the alert stock and the quantity
requested. It then has the same advantages and is adapted to the sporadic expressions of demand
57
that can be seen during commercial campaigns. This makes it easier to immediately replenish
the stock in the event of aberrant flows.
Thus, for the packaging management system chosen, we prioritize the S,R policy.
Alert Stock (𝑆𝑎 ) = 𝑚 × 𝑑 where d is the delivery time in days and m is the average
consumption of the reference to be supplied per week. The delivery time is defined by the time
it takes to know the stock level, the validation time of the warehouse manager, the time it takes
for the inventory manager to make it available and the time it takes to update the stock. Lots of
variants that delay supply.
Replenishment level (𝑁𝑟 ) = (𝑚 × 𝑇) + 𝑆𝑎 where T is the fixed replenishment time, in this case
one (01) week.
Monthly Consumption
Description
January February March April
Lot de 25 Sachets Orange imprimé Jumia 64.5 Cm x 52 Cm - Orange 150 155 208 195
Source : L’étudiant
The time to know the stock level is zero because the safety stock is deviated from the regular
consumption quantity. Each time this quantity is exhausted, the need for an order merges with
the expression of this demand. The average validation time of the Warehouse Manager is two
58
(02) hours. This long delay is due to the workload of the latter. The availability time by the
general stock manager is also two (02) hours for the same reasons. The time to update the stock
is thirty (30) minutes. It consists of updating the alert stock and then counting. In order to
compensate for any eventuality, we add a thirty (30) minute deviation-error.
The delivery time (d) is therefore five (05) hours or 0.2 days.
Since the unit of time is the month, we apply the following moving average algebra to the
𝑚𝑜𝑦𝑒𝑛𝑛𝑒 𝑑𝑒 𝑐𝑜𝑛𝑠𝑜𝑚𝑚𝑎𝑡𝑖𝑜𝑛 𝑑𝑒𝑠 3 𝑑𝑒𝑟𝑛𝑖𝑒𝑟𝑠 𝑚𝑜𝑖𝑠⁄
volume for the month of May : ( 4) . This
is in order to reduce the unit of time to the week and take into account the different variations
in demand over the previous three (03) months.
𝐻𝑜𝑤𝑒𝑣𝑒𝑟, 𝑡ℎ𝑒 𝑙𝑒𝑣𝑒𝑙 𝑜𝑓 𝑡ℎ𝑒 𝑎𝑙𝑒𝑟𝑡 𝑠𝑡𝑜𝑐𝑘 (𝑆𝑎 ) is given in the table below :
Monthly Consumption 𝑆𝑎
Description
January February March April Mai
Lot de 25 Sachets Orange imprimé Jumia 64.5 Cm x 52 Cm – Orange 150 155 208 195 10
Source : L’étudiant
Similarly, we determine the level of replenishment that corresponds to the maximum level to
be filled for each reference. Since the alert stock has already been defined for each SKU, let's
trivially determine the average weekly consumption of the SKUs to be supplied. This level (m)
is obtained by the following formula :
replenishment level.
59
Table 15 Determination of the level of replenishment by reference
Monthly Consimption 𝑆𝑎 m 𝑵𝒓
Description
January February March April Mai Mai Mai
Source : L’étudiant
Thus, the level of replenishment, the variable weekly consumption, adjustable according to the
flow of the previous months and the alert stock are defined for all references. This simple model
should no longer allow for stock-outs.
2.2.2 Recommandations
The recommendations presented relate to the causes identified by the Ishikawa diagram
affecting the packaging management system. At each M, we present some recommendations to
increase the performance of the order preparation process.
Method
- Maintain the S,R packaging management policy;
- Adapt the alert stock and the average weekly consumption in the event of an abnormal
flow;
- Carry out an inventory every 48 hours to ensure that the theoretical stock corresponds
to the physical stock.
Material :
- Identify storage bins according to references in order to facilitate inventories and service
on packing lines;
60
- Set up a G-Sheets or Excel file in order to share the data with the people directly
involved in order preparation (Team Leader, Warehouse Manager, Inventory Manager,
etc.).
Manpower :
- Perform ongoing training to agents on Ms Excel and G-Sheets data management tools;
- Empower teams and instill a sense of purpose.
Milieu :
- Improve collaborative exchanges between sales representatives and order pickers in
order to be notified of all commercial campaigns and the level of flow accordingly.
These recommendations may eventually help to improve working conditions, as operators and
team leaders will be motivated by the common goal of drastically reducing the company's
operating costs.
61
PART TWO CONCLUSION
The second part reviewed the two main causes that affected the level of service of order picking.
These include the inaccuracy of order forecasts and the non-existent packaging management
model; To which we have responded with an exponential smoothing forecasting model and a
packaging management policy by reorder point and replenishment level. The implementation
of these policies will have to be done again after a precise and short training of the agents,
which should not normally lead to any major change management.
62
GENERAL CONCLUSION
As Jumia Côte d'Ivoire strives for profitability, the company has set itself the goal of drastically
reducing the costs generated by its business. These costs, essentially made up of operating and
labor costs, are generated by the operations of its distribution center, the only one used to serve
the entire national territory for orders at outlets or direct to the consumer.
This being the case, the aim of our study could not escape the need to rationalize the costs
generated by order preparation operations. Among these, the management model for order
packaging and the number of operatives required on a daily basis were deemed to be cost-
intensive, given the volatility of the daily order flow.
As a result, at the start of the study, a focus group with team leaders led to the conclusion that
imprecise order forecasts, the absence of a packaging management model and poorly applied
processes were preventing the various order-picking operations from being perfected.
Reviewing the existing processes gave us a better understanding of the issues at stake.
In order to remedy the situation, we began by using the QQOQCP method and the Ishikawa
diagram to identify the root causes of the under-performance. We then proposed a new
forecasting model, exponential smoothing, which we considered more relevant, followed by a
packaging management policy based on replenishment by order point and replenishment level.
Recommendations have been made to continually improve these new processes.
People management also remains at the heart of improved performance, given that operations
are still less mechanized. A better understanding of the management of the various people
involved in order preparation is therefore essential, in order to considerably reduce error rates
and the deployment of return logistics.
Improving the performance of the order-picking department depends entirely on the application
of the measures mentioned above.
63
BIBLIOGRAPHY
Books
Memoirs
64
- Joseph FARDON (2022), « Planification des approvisionnements des produits de
promotion dans une entreprise de Grande distribution : Cas de CFAO Retail CI » (81
pages) ;
- Schontchon Chigata Emmanuel COULIBALY (2020), « Amélioration du processus de
traitement de commandes dans une entreprise de e-commerce : cas de Jumia CI » (81
pages)
65
WEBOGRAPHY
- EDGAR Entity Landing Page (sec.gov) consulté le 02/03/2023
- Jumia Reports First Quarter 2023 Results (yahoo.com) consulté le 03/04/2023
- Emballage e-commerce : guide pratique 2023 (dssmith.com) consulté le 02/06/2023
- Qu'est-ce que le Fulfullment ? Explication du terme (1min30.com) consulté 03/06/2023
-
66
ANNEXES
Annexe 1 : Financial Statements 2023 (Jumia Technologies AG | Financials & Filings)
67
Annexe 2 : Dashboard KPI following up
68
Annexe 3 : Interface de OMS
69
TABLE DES MATIERES
SOMMAIRE ................................................................................... Erreur ! Signet non défini.
DEDICATION ........................................................................................................................... 2
THANKS .................................................................................................................................... 3
FOREWORD ............................................................................................................................. 4
GLOSSARY ............................................................................................................................... 9
RÉSUMÉ.................................................................................................................................. 11
70
2.2 Analytical Methods ................................................................................................ 35
BIBLIOGRAPHY .................................................................................................................... 64
WEBOGRAPHY ...................................................................................................................... 66
ANNEXES ............................................................................................................................... 67
71