A Big Data Analytics Architecture Framework for the Production and International Trade of Oilseeds and Textiles in Sub-Saharan Africa (SSA)

Abstract

Among the most revolutionary technologies are big data analytics, artificial intelligence (AI) and robotics, machine learning (ML), cybersecurity, blockchain technology, and cloud computing. The project was focused on how to create a Big Data Analytics Architecture Framework to increase the production capability and global trade for Sub-Saharan Africa’s oilseeds and textile industries (SSA). The infrastructure, e-commerce, and disruptive technologies in the oilseeds and textile industries, as well as global e-commerce, all demand large investments. The pragmatic paradigm served as the foundation for the research approach. This study employed a review of the literature, document analysis, and focus groups. For the oilseeds and textile sectors in SSA, a Big Data analytics architectural framework was created. The Hadoop platform was created as a framework for big data analytics. The open-source Hadoop platform offers the analytical tools and computing capacity needed to handle such massive data volumes. It supports E-commerce and is based on the Hadoop platform, which offers the analytical tools and computing power needed to handle such massive data volumes. The low rate of return on investments made in breeding, seed production, processing, and marketing limits the competitiveness of the oil crop or legume seed markets.

1. Introduction

Massive amounts of data are produced in the Internet of Things (IoT) age from a number of heterogeneous sources, such as mobile devices, sensors, and social media. Among the most revolutionary technologies are big data analytics, artificial intelligence (AI) and robotics, machine learning (ML), cybersecurity, blockchain technology, and cloud computing. The two basic features of machine learning are the automatic analysis of large data sets and the creation of models for the broad relationships between data (ML). Analyzing large amounts of data to find information—such as hidden patterns, correlations, market trends, and customer preferences—that can assist organizations in making strategic business decisions is known as big data analytics [1, 2]. Volume, value, variety, velocity, and veracity are the five characteristics that define Big Data, as shown in Figure 1.

Figure 1.

Big data characteristics.

Legumes, shea butter, groundnuts, and soybeans are significant crops in Sub-Saharan Africa (SSA) because they offer a range of advantages in terms of the economy, society, and the environment. Sub-Saharan Africa contributes a relatively little amount to global agricultural output despite having over 13% of the world’s population and about 20% of its land area being used for agriculture, claims the [3]. The research paper is purposed to develop a Big Data Analytics Architecture Framework for the Production and International Trade of Oilseeds and Textiles in Sub-Saharan Africa (SSA).

1.1 Background

More over 950 million people live in Sub-Saharan Africa (SSA), accounting for roughly 13% of the global population. Oilseed production in SSA is expected to increase by 2.3 percent per year to 11 Mt. by 2025, accounting for barely 2% of global production.

Although expected increase in Southern Africa is more modest at 16 percent, the base is significantly greater, and Southern Africa accounts for the largest proportion of additional protein meal use in absolute volumes. Southern (1.4 percent per year) and Eastern Africa (1.2 percent per year) are expected to grow at the quickest rates to 2025. Protein meal use is increasing across most of SSA as livestock industries strengthen in the future years, with Western Africa (43 percent) and Eastern Africa (43 percent) seeing the largest rise (32 percent). Oilseed production in SSA is expected to increase by 2.3 percent per year to 11 Mt. by 2025, accounting for barely 2% of global production. Nonetheless, total imports into SSA are expected to grow at a 3.7 percent annual rate, with Nigeria (4 percent per year), Sudan (5 percent per year), Ethiopia (6 percent per year), and Kenya (3 percent per year) accounting for the majority. Per capita consumption has grown at a rate of 2.1 percent per year, making it one of the fastest growing commodities in the region during the last decade. Over the next decade, Sub-Saharan Africa’s net food imports are expected to rise, however productivity-boosting investments could counteract this trend. Despite the fact that agricultural productivity has increased significantly over the last decade, SSA remains the world’s most food insecure region, with inconsistent progress toward hunger eradication. The world oilseeds supply and distribution in million metric tons for the period 2017 to 2022 is shown on Table 1.

	2017/18	2018/19	2019/20	2020/21	2021/22
Production
Copra	5.78	5.82	5.7	5.59	5.86
Cottonseed	45.25	42.97	43.55	40.81	42.75
Palm Kernel	18.69	19.46	19.32	19.03	20.05
Peanut	47.15	46.71	48.14	50.25	50.29
Rapeseed	75.28	72.85	69.6	73.59	71.18
Soybean	343.74	362.44	340.15	368.12	349.37
Sunflowerseed	48.01	50.66	54.2	49.25	57.38
TOTAL	583.9	600.91	580.65	606.64	596.87
Imports
Copra	0.13	0.2	0.15	0.08	0.08
Cottonseed	0.87	0.73	0.81	0.83	0.97
Palm Kernel	0.18	0.16	0.14	0.15	0.17
Peanut	3.08	3.53	4.34	4.31	4
Rapeseed	15.72	14.64	15.71	16.66	13.97
Soybean	154.11	146.02	165.12	165.47	154.46
Sunflowerseed	2.38	2.89	3.34	2.73	2.2
TOTAL	176.47	168.17	189.61	190.24	175.86
Exports
Copra	0.16	0.18	0.28	0.1	0.13
Cottonseed	0.89	0.84	0.88	0.96	1.16
Palm Kernel	0.16	0.07	0.08	0.06	0.05
Peanut	3.51	3.83	4.95	4.89	4.64
Rapeseed	16.53	14.62	15.92	17.98	13.84
Soybean	153.27	148.97	165.21	164.51	155.57
Sunflowerseed	2.75	3.24	3.66	2.91	2.59
TOTAL	177.28	171.75	190.97	191.41	177.97
Crush
Copra	5.67	5.83	5.56	5.52	5.71
Cottonseed	33.73	32.75	33.62	31.95	33.2
Palm Kernel	18.62	19.42	19.29	19.01	20.08
Peanut	18.15	18.05	19.24	19.86	20.1
Rapeseed	68.45	68.03	68.41	71.45	70.2
Soybean	295.44	298.61	312.31	315.08	313.68
Sunflowerseed	44.17	46.52	49.31	45.13	47.34
TOTAL	484.24	489.2	507.73	508	510.31
Ending Stocks
Copra	0.12	0.1	0.05	0.05	0.07
Cottonseed	1.96	1.82	1.61	1.41	1.42
Palm Kernel	0.23	0.26	0.24	0.25	0.23
Peanut	5.16	5.08	4.67	4.89	4.33
Rapeseed	8.14	9.93	7.81	5.96	4.27
Soybean	99.84	114.19	94.66	99.91	85.24
Sunflowerseed	2.79	2.57	2.92	2.56	7.61
TOTAL	118.24	133.95	111.96	115.02	103.16

Table 1.

Major oilseeds world supply and distribution (2017–2022) [million metric tons].

Source: https://apps.fas.usda.gov/psdonline/circulars/oilseeds.pdf

The world production of oilseeds for the period 2017–2022 is shown on Figure 2.

Figure 2.

World production of oilseeds (2017–2022).

The world oilseeds crush distribution for the period 2017–2022 is shown on Figure 3.

Figure 3.

World oilseeds crust distribution (2017–2022).

The focus of the researchers was on how to use and implement Big Data to improve production for both oilseeds and textile production and international trade for Sub-Saharan Africa (SSA).

The top 15 textile exporters in Sub-Saharan Africa (SSA) are shown on Table 2 below and illustrated on Figure 4.

	1995	2000	2005	2010	2015	2016
Lesotho		146365.92	408337.98	293625.99	310412.35	304867.13
Kenya	40557.59	46921.64	286480.04	212267.49	381118.11	352218.08
Mauritius	201,844	259,609	175787.13	127105.49	221933.63	203340.45
Madagascar	7475.2	115429.39	293757.75	58139.23	54429.66	108345.99
South Africa	164868.09	187000.1	107985.72	23786.08	26942.7	25108.16
Swaziland		33407.42	168769.77	97887.4	2807.2	1067.87
Tanzania	6084.74	253.87	4437.83	2159.59	27999.56	37883.39
Botswana		9028.59	31459.14	12209.52	8685.86	4981.05
Ethiopia(excl. Eritrea)	971.4	30.98	3829.68	7113.17	18799.72	34457.11
Namibia		196.09	56050.93	47.06	230	122.43
Malawi	2509.89	7653.83	24018.24	10728.07	6437.02	1603.53
Zimbabwe	15484.16	21574.02	3086.21	87.37	130.48	99.08
Ghana	3216.37	718.84	5749.01	1071.03	9620.28	6631.52
Cameroon		2769.28	407.24	749.97	1003.44	342.41
Uganda		5.07	5143.94	461.64	73.47	78.62

Table 2.

Top 15 SSA exporters of textiles and clothing to US (US$‘000).

Source: World Bank.

Figure 4.

The top 15 SSA exporters of textiles and clothing to US (US$‘000).

Many textile and apparel inputs now produced in SSA nations can be made more competitive by new or increased investment or other methods, especially as output of these inputs is restricted and diminishing in many cases. New or expanded investment, as well as other initiatives, could help the industry maintain or expand present production and export levels of these inputs, as well as extend the possibility for new product development.

This paper aims to develop a Big Data Architecture framework for oilseeds and textile industry production and international trade for SSA.

1.2 Statement of the problem

Organizations struggle to manage and track the growth of both new and old open-source big-data solutions, which are continually expanding. The considerable volume of data produced by a wide range of sources, including as information services, Internet of Things (IoT) devices, social media, and mobile devices, is not only too large but also moves too quickly and is too complex to be handled and stored by conventional techniques. The sector is driven by the data’s exponential growth, which also draws researchers to create new models and scalable methods for handling big data. A well-known open-source framework for big-data analytics, Apache Hadoop is made to integrate with a number of other open-source technologies to allow for the storing and processing of large amounts of data using commodity hardware clusters. A distributed file system, cluster administration, storage, distributed processing, programming, data analysis, data governance, and data pre-processing tools are all included in the Hadoop Stack. The production and global commerce of oilseeds and the textile industries should take this into account.

The African Growth and Opportunity Act (AGOA), a non-reciprocal trade preference program, was established by the US Congress in 2000 to assist developing SSA nations in improving their economies through increased exports to the US. Notably, the “third-country fabric clause” in AGOA permits US clothing imports from specific SSA nations to qualify for duty-free treatment even if the clothing products use yarns and fabrics manufactured by non-AGOA members, such as China, South Korea, and Taiwan. Furthermore, AGOA trade preferences offer much bigger duty savings for manmade-fiber products, which are subject to higher U.S. tariffs, even though SSA nations generate largely cotton-based textile and garment inputs due to a plentiful availability of local cotton. Cotton yarn, cotton knit fabric, denim fabric, and to a lesser extent cotton woven shirting fabric appear to have the most potential for competitive production in SSA countries, either for direct export to or use in downstream apparel production for export to the United States, the EU, and similar markets. However, because the manmade-fiber textile and apparel sectors are underdeveloped in most SSA countries, it is not possible to produce these products. All of these products may be competitive in some local and regional markets because numerous SSA industry sources reported producing textile and garment inputs for both regional consumption and export beyond the region.

1.3 Research aim

This paper aims to develop a Big Data Architecture framework for oilseeds and textile industry production and international trade for SSA.

1.4 Research objectives

The objectives of the research include the following:

1. To identify areas of Big Data applications that can help the oilseeds and textile industries in SSA increase their production and worldwide commerce.

2. To develop a Big Data Analytics architecture framework for usage by organizations in the oilseeds and textile production industry, as well as international trade in SSA.

3. To determine the competitive challenges facing Sub-Saharan Africa (SSA) in the production of oilseeds and textile industry.

4. To evaluate yield production capacity and competitive variables across SSA for both oilseeds and textile industry.

1.5 Research questions

1. What are some Big Data applications that can help SSA boost its oilseed and textile output and international trade?

2. How do you create a Big Data Analytics architecture framework to assist and improve oilseed, textile, and international trade production?

3. What are the competitive issues in the oilseed and textile industries in Sub-Saharan Africa?

4. What has been the state of production capacity and competitive factors in the oilseeds and textile industries across SSA?

2. Critical context

2.1 Literature review

Big Data (Data Intensive) Technologies aim to process (1) highvolume, highvelocity, high-variety data (sets/assets) to extract intended data value and ensure high-veracity of original data and obtained information; this calls for cost-effective, innovative forms of data and information processing (analytics) for improved insight, decision-making, and process control; all of these call for (should be supported by) new data models (supporting all data states and stages during the whole data lifecyle) and infrastructure services and tools. Generally, the term “big data” refers to the rapidly expanding volume and velocity of data sets that are being accessible and connected. According to studies, big data may generally be defined using the four (4) V’s of big data. The five properties of volume, value, diversity, velocity, and veracity are frequently used to describe big data, which is a collection of data from various sources. Big data analytics, which some academics define as the capacity to compile and analyze those fine-grained data sets, is already altering how insurers see sizable client bases, manage risks, and meet the diverse needs of their clients. Kabanda [4] defines big data analytics as the straightforward application of analytics approaches to significant data sets. The five properties of volume, value, diversity, velocity, and veracity are frequently used to describe big data, which is a collection of data from various sources. Many significant businesses employ software for machine learning, artificial intelligence, data mining, cybersecurity, and other big data. Big data analytics, which some academics define as the capacity to compile and analyze those fine-grained data sets, is already altering how insurers see sizable client bases, manage risks, and meet the diverse needs of their clients. OECD-FAO [4] defines big data analytics as the straightforward application of analytics approaches to significant data sets.

The types of analytics applicable in the oilseeds and textile industries are shown on Figure 5 and are briefly explained below:

1. Descriptive analytics - Descriptive analytics aims at describing and analyzing historical data collected on students, teaching, research, policies and other administrative processes. The goal is to identify patterns from samples to report on current trends.

2. Predictive analytics - Predictive analytics can provide organizations with better decisions and actionable insights based on data. Predictive analytics aims at estimating likelihood of future events by looking into trends and identifying associations about related issues and identifying any risks or opportunities in the future. Predictive analytics could reveal hidden relationships in data that might not be apparent with descriptive models, such as demographics, etc.

3. Prescriptive analytics - Prescriptive analytics helps organizations assess their current situation and make informed choices on alternative course of events based on valid and consistent predictions. It combines analytical outcomes from both descriptive and predictive models to look at assessing and determining new ways to operate to achieve desirable outcomes while balancing constraints indicated that prescriptive analytics enables decision makers to look into the future of their mission critical processes and see the opportunities as well as presents the best course of action to take advantage of that foresight in a timely manner.

Figure 5.

Types of analytics.

Machine learning is a method for instructing computers to learn (ML). Big data analytics is known to be automated using machine learning, which also creates models of the fundamental relationships in the data. The way we teach, learn, and study in the educational setting could be completely changed by machine learning (ML). Localization, transcription, text-to-speech, and personalisation are just a few of the ways that machine learning is expanding the reach and impact of online learning content [5]. Data mining can be handled through machine learning. According to Truong [6], there are three types of ML:

1. Supervised learning: where training examples are given to the methods in the form of inputs labeled with corresponding outputs;

2. Unsupervised learning: where unlabeled inputs are given to the methods;

3. Reinforcement learning: where data used is in the form of sequences of actions, observations, and rewards.

Machine Learning essentially includes programming analytical model construction and is a technique of big data analytics [7].

Data mining is the process of discovering anomalies, trends, and correlations in large data sets in order to predict outcomes [8]. Data mining is most usually characterized as the process of searching massive sets of data for patterns and trends using computers and automation, then translating those findings into business insights and predictions. Data mining is an important element of data analytics and one of the fundamental disciplines in data science, in which advanced analytics techniques are used to extract meaningful information from large data sets. While both are valuable for spotting patterns in enormous data sets, they work in quite different ways. The practice of detecting patterns in data is known as data mining. And, while data mining is sometimes used as part of the machine learning process, it does not necessitate continual human engagement (e.g., a self-driving car relies on data mining to determine where to stop, accelerate, and turn).

Computer systems that imitate human intellectual processes, such as learning, reasoning, and self-correction, are referred to as artificial intelligence (AI). The ability of AI to arrive at a solution based on facts rather than a predetermined series of procedures is what most closely mimics the human brain’s thinking function. Artificial intelligence (AI) is defined by its ability to replicate human behavior and cognitive processes, to capture and preserve human expertise, to respond swiftly, and to manage large amounts of data quickly.

As a result, cyber security has become an important concept in everyday life, and cyber security knowledge is critical in preventing cyber attacks on people and systems. With the rise of a global and borderless information culture, the internet has brought and continues to present new opportunities to all countries globally, as technologies play a key role in social and economic development [9]. With the rise of a global and borderless information culture, the internet has brought and continues to present new opportunities to all countries globally, as technologies play a key role in social and economic development [9]. Cyber security refers to strategies used to secure sensitive data, computer systems, networks, and software applications from cyberattacks, according to [10]. The cyber security concept’s main purpose is to protect data confidentiality and integrity while also providing data availability when it’s needed. However, as the nature of cyber threats changes, so does public concern about cyber security issues like social engineering and phishing.

The foundation of Big Data architecture is infrastructure. In every Big Data project, having the correct tools for storing, processing, and analyzing your data is critical [11].

3. Oilseeds and textile production competitive challenges in SSA

Certain competitive challenges affected nearly all the SSA countries, as described below.

• Insufficient demand from the apparel sector

A healthy and thriving garment industry offers the stable market demand for textile and apparel inputs that is required to support capital expenditures that take longer to recover than apparel investments.

• Lack of knowledge of regional and international market opportunities

Many industry experts pointed to a lack of marketing and business contacts, both within the SSA region and in international markets. According to industry sources, the USAID has aided in the development of regional and international market potential, but further assistance is needed.

• Insufficient supply of reliable electricity at competitive rates

Many nations in the SSA region have among of the highest electricity tariffs in the world, and many countries have an unstable electrical supply, which adds to producers’ expenses. Electricity outages also diminish efficiencies and lower quality in yarn and fabric production.

• Insufficient supply of clean water and wastewater treatment facilities

Many countries lack access to clean water, which is required for the manufacturing of yarn and fabric, particularly for finishing and dyeing activities. Intraregional trade is further hampered by a lack of adequate transportation networks within SSA.

• Lack of access to capital at competitive rates

When capital is available, the high cost of capital not only discourages new investment in yarn, fabric, and other inputs, but it also raises the costs of existing production. The finished products created on this machinery, particularly woven textiles, are often not of adequate quality for export to the United States, the EU, or similar markets, or for use in downstream commercial garment production for export to these countries.

• Scarcity of trained/skilled labour

According to industry sources, there is a shortage of trained workers in the textile and garment industries, particularly in nations without a substantial manufacturing base.

4. Conceptual framework on adoption of big data analytics

The study is guided by the conceptual framework shown on Figure 6 below:

Figure 6.

Conceptual framework.

4.1 Research methodology

The study’s research philosophy, as well as the research design, research approach, data collection instruments, target population, sampling method, and data processing techniques, are all explained in the Research Methodology. The research philosophy, approach, strategy, choice, time horizon, and techniques and processes constitute the layers, as shown on Figure 7.

Figure 7.

Research onion (Saunders et al., 2009:138).

The Mixed Method Research and the Pragmatism paradigm utilized in this study are closely related on a philosophical level (MMR). A worldview or paradigm known as pragmatism ought to guide the majority of mixed-methods studies. It is a problem-focused attitude that holds that the best research techniques are those that contribute most significantly to the solution of the research topic. When conducting social science research, this frequently entails combining quantitative and qualitative methodologies to assess various facets of a research subject. The pragmatic worldview served as the foundation for the Mixed Methods Research technique. A mixed-methods strategy was used in this study, combining qualitative (Focus Group discussions) and quantitative techniques (a questionnaire). System logs, document analysis, and a literature review were also utilized in this study.

The purpose for the Focus Group discussion was to research and determine on how to use and implement Big Data to improve production for both oilseeds and textile production and international trade for Sub-Saharan Africa (SSA). The Focus Groups were derived from 10 Groups of Masters students at the University of Zimbabwe in the 2021 cohort who then were tasked to conduct surveys and interview the management of various corporates in Zimbabwe and other nearby Southern African countries involved in the oilseeds and textile industries. Secondary data was collected form the World Bank, FAO [FAOSTAT, www.faostat.org] and US Department of Agriculture (https://apps.fas.usda.gov/psdonline/circulars/oilseeds.pdf) for analysis.

5. Results and analysis

5.1 Critical challenges around the application of big data

5.1.1 Analytics in the oilseeds and textile industries

The Critical challenges around the application of big data analytics in the oilseeds and textile industries as obtained from the participants of the focus group are:

• Lack of Talent - While there is a massive demand for experienced data experts, there simply is not enough supply. Unfortunately, this deficit has not yet been addressed by most top universities as data science programs are still lacking.

• Storage and Scalability Issues - The volume of data being generated exceeds the processing power of currently accessible Big Data tools. This can cause significant issues and force systems to crash or slow down, leading to a negative experience and a reduced quality of the analysis.

• Security - Security protocols were not built for a Big Data world and need to be reworked to account for the volume of data that Big Data uses in its analysis. The cybersecurity systems in most of the organizations in the oilseeds and textile industries is still in their infancy stages of development.

5.1.2 Implementation of an AI Chatbot and E-commerce

Huge investments are required in the infrastructure that is inclusive of AI Chatbots and E-Commerce for all the corporates and entities involved in the oilseeds and textile business. Machine learning is a key feature of AI chatbots since it allows them to learn and improve based on their experiences. Electronic business can allow an organization to implement cybercash, Electronic Data Interchange (IDE), electronic advertising, business to business and business to customer online transactions on a worldwide scale. Small businesses can compete with well-established and capital-rich businesses on a global level thanks to electronic commerce and sound strategy and policy methods. From a business process standpoint, electronic business is the application of technology to the automation of company transactions and workflow. Electronic trading of goods and services, on-line delivery of digital content, electronic fund transfers, electronic share trading, electronic bills of lading, commercial auctions, collaborative design and engineering, on-line sourcing, public procurement, direct consumer marketing, and after-sales services are all examples of transactions in the global information economy.

The eight Unique Features of E-commerce Technology required includes the following:

1. Ubiquity

2. Global reach

3. Universal standards

4. Information richness

5. Interactivity

6. Information density

7. Personalization/customization

8. Social technology

Each of the corporates involved in the oilseeds and textile industries is encouraged to invest extensively in AI Chatbots and E-Commerce in order to build a basis upon which to successfully implement a Big Data Analytics Framework.

5.1.3 Data analysis of oilseeds production in SSA

1. Soybean

   Soybean is a vital crop for at least one million African smallholder farmers. Other factors have contributed to rising soybean demand, such as the need for domestic processing to meet rising domestic demand for soybean meal, especially for the poultry feed industry, and the favorable outlook for edible oil. The United States, Brazil, and Argentina, as the world’s top three soybean exporters, will continue to account for approximately 90% of global soybean, soybean meal, and soybean oil exports in the coming decade. The soybean production levels and yield per hectare for the world, Sub-Saharan Africa (SSA) and other countries are shown on Figure 8 below. Both area expansion and yield growth have contributed about equal amounts to the reported growth in soybean output in SSA, with yearly growth rates of 3% in area and 3.5 percent in yield.

2. Groundnuts

   After oil palm, soybean, rapeseed, and sunflower, groundnut is the world’s fifth most important oilseed crop. Groundnut is a major oil, food, and feed legume crop that is produced in over 100 countries, covering 25.44 million hectares and producing 45.22 million tons of pods in 2013. Despite Africa’s declining share of the global groundnut market, the crop still accounts for a large portion of export revenues in several countries (for example, 8% in Senegal and over 84 percent in Gambia in 2002). Groundnut is a nutritious food that helps to improve the health of rural people. Groundnut haulms, which contain 8–15 percent protein, 1–3 percent fats, 9–17 percent minerals, and 38–45 percent carbs, are used as cattle feed in both fresh and dried form, as well as for making hay and silage.

3. Cowpea

   About 95 percent of global cowpea production is produced in Sub-Saharan Africa (SSA), with West Africa producing over 80 percent of Africa’s share. Over 65 percent of cowpea is produced by poor households in Nigeria, meaning that cowpea is primarily produced by the poor, who stand to benefit from cowpea research and extension. The basic analysis shows the production of cowpeas and yield production per ha shown on Figure 9.

   Despite these encouraging signs, cowpea yields remain low due to a variety of production restrictions as well as a lack of adoption of improved varieties and agronomic approaches.

4. Shea Butter

   Shea grows over an estimated 1 million km² between western Senegal and northwestern Uganda in the Sudan zone’s dry savannas, woods, and parklands. According to OECD-FAO [3], Nigeria has the greatest potential for shea nut production, with high production zones in Benin, Burkina Faso, Cote D’Ivoire, Ghana, Mali, and Nigeria. Because producers, particularly women, and the private sector in nations where production capacity is not fully utilized, the potential of production capacity is not fully realized. Because producers, particularly women and the private sector in countries where shea trees grow, are not completely participating in the value addition sales of the nuts or butter, the potential of the production capacity is not fully realized. When Ghana’s shea production potential is completely realized, this amount is predicted to quadruple.

Figure 8.

Soybean world production and yield/ha.

Figure 9.

SSA cowpea production and yield/ha.

6. The key recommendations for the oilseeds and textile industries is the need to attend to the problems for each of the following areas

1. Poor access to improved seeds

   A multitude of issues contribute to the existing seed system’s failure to offer improved varieties of oil crop and legume seeds to smallholder farmers. Many farmers have grown accustomed to receiving free seed from non-governmental organizations (NGOs) and are unappreciative of the investment necessary. Based on lessons learned in other regions of the world, establishing a Foundation Seed Enterprise committed to the production and distribution of foundation/basic seed can aid seed companies interested in commercializing enhanced publically developed varieties.

2. Lack of farm machinery

   Despite the development of yield-enhancing technologies over the last three decades, labor-intensive farming practices continue to prevail, and crop products are still processed manually at home. However, due to poor input marketing arrangements, inorganic amendments are rarely available in cheap quantities to farmers.

3. Low soil fertility

   However, due to poor input marketing arrangements, inorganic amendments are rarely available in cheap quantities to farmers.

4. Input market constraints

   Improved seed, fertilizers, crop protection products, and novel agronomic practices are all examples of science and technology that can help accelerate agricultural growth. Because of a lack of oversight, pesticides that are old or contaminated are widely used.

5. Output market constraints

   Because of a number of structural and institutional obstacles that impede market participation, smallholders have not been able to respond effectively to potential soybean market possibilities. Low quality grain, insufficient supply, and high cleaning costs restrain processors and traders, whilst market intermediaries suffer high assembly costs, high market risk, and cash flow issues. Improving smallholder farmers’ market access and competitiveness would necessitate new types of market institutions that allow contract formulation and enforcement, as well as vertical and horizontal coordination of production and marketing tasks. Farmers’ awareness and access to new information, expected benefits and local availability of new technologies, market access and opportunities, and access to credit and other policies that enable farmer investment in new technologies have all been shown to be major drivers of research product dissemination and adoption.

6. Low levels of technology adoption

   A number of socioeconomic and targeting studies (http://www.icrisat.org/impi-tl-2.htm) demonstrate that new variety acceptance has been slow and sluggish, with old varieties launched 15–20 years ago still occupying much of the production area. Farmers’ awareness and access to new information, expected benefits and local availability of new technologies, market access and opportunities, and access to credit and other policies that enable farmer investment in new technologies have all been shown to be major drivers of research product dissemination and adoption.

7. Textile production capacity and competitive factors

The textile production capacity and competitive factors for each top country is summarized on Table 3, indicating the competitive advantages and disadvantages of each country.

Group 1 Countries	Textile and apparel input production	Competitive factors
Ethiopia	The Ethiopian textile sector includes eight vertically integrated textile mills, along with stand-alone spinning mills for yarn and thread production. Most of the yarn spun in Ethiopia is used in the production of woven cotton fabric. In addition to cotton yarn and woven fabric, Ethiopia’s textile sector also produces acrylic yarn, nylon fabric, woolen and waste-cotton blankets, bedsheets, and sewing thread. Ethiopia currently produces cotton and silk yarn for domestic hand-loomed production of niche products, such as home furnishings, for export to the United States, Canada, and Europe.	Competitive advantages: • large potential domestic apparel market; • domestic production of raw materials (cotton, silk); • stable political and business environment; • access to Ethiopian government-supported investment incentives and financial assistance; Competitive disadvantages: • import competition from used clothing; • low cotton production; cotton contamination; • poor transportation infrastructure • underutilized industrial capacity; • outdated machinery and equipment; • low labor productivity; • lack of skilled labor
Kenya	The Kenyan textile industry has contracted since the 1990s and currently consists of three vertically integrated firms and a few smaller, nonintegrated firms. Kenya’s vertically integrated firms produce cotton (including organic) and synthetic yarn, and knitted and woven fabric for use in apparel exported to the United States and the EU. Some yarn and fabric is also sold regionally.	Competitive advantages: • export-oriented apparel industry; • relatively skilled labor; Competitive disadvantages: • poor roads; • high-cost electricity; • limited and high cost of financing for new equipment
Lesotho	Lesotho has one vertically integrated denim textile mill that spins cotton yarn, dyes the yarn, weaves the fabric, and cuts and sews the finished denim jeans. The mill reportedly produces 10,800 tons of openended ring-spun cotton yarn, and 18 million yards of denim fabric a year for regional apparel manufacturers producing for the export market. Lesotho primarily exports woven fabric to other apparel-producing African countries. The vast majority of Lesotho’s apparel exports are to the U.S. market.	Competitive advantages: • export-oriented apparel industry; • government investment support for plant acquisitions; Competitive disadvantages: • poor water/wastewater and internal transport infrastructure; • low labor productivity; • high HIV/AIDS prevalence rates; • lack of skilled labor
Madagascar	The Malagasy textile industry consists of one large vertically integrated woven textile and apparel firm that consumes most of its own fabric production, two small knit apparel firms that produce their own knit fabric, and another firm that weaves fabric for blankets. The Malagasy apparel sector is geared to supply the U.S. and EU markets.	Competitive advantages: • export-oriented apparel industry; • availability of skilled and productive labor; • government investment incentives and support; Competitive disadvantages: • diminishing supply of domestic cotton; • high-cost, unreliable electricity; • political instability; • high cost of capital; • poor road infrastructure
Mauritius	The Mauritian industry is concentrated among 10 large textile and apparel groups that collectively account for 75 percent of total textile and apparel exports. The textile and apparel input industry in Mauritius produces yarn and knit fabric mostly for vertical operations, but also for local and regional apparel manufacturers. Mauritius exports textile and apparel inputs to the region and finished apparel primarily to the EU.	Competitive advantages: • export-oriented apparel industry; • market linkages with EU apparel buyers; • favorable business environment; • government support in product and market diversification; • relatively modern machinery and equipment; • shorter lead times to the region and to some EU customers; • availability of skilled labor; Competitive disadvantages: • small domestic apparel market; • increased labor costs due to labor shortages; • long lead times to the United States and to some EU customers; • increasing land and energy costs; • additional costs associated with geographic isolation
Nigeria	The Nigerian textile industry has contracted since the 1990s and currently consists of 20 or fewer factories. Some larger textile firms are vertically integrated from cotton ginning to spinning, weaving, dyeing, printing, and finishing. The major textile firms produce a variety of products, including polyester staple fiber and filament, yarn, greige cloth, and wax prints. Nigerian printed fabric is sold as loose cloth, rolls, or pieces to the domestic market. Nigerian textile exports are focused on the EU market.	Competitive advantages: • large potential domestic apparel market; • history of cotton and integrated textile production; • availability of skilled labor; Competitive disadvantages: • lack of a developed apparel industry; • increased import foreign competition (ethnic cloth and used clothing); • cotton quality issues; • poor infrastructure, particularly electricity
South Africa	The South African textile sector is relatively large and encompasses the full range of manufacturing operations, including production of fiber, thread, yarn, knit and woven fabric, nonwovens, trim and accessories, and dyeing and finishing operations. There are currently 11 firms in South Africa producing yarn. Five firms manufacture nonwovens, and reportedly seven firms produce trim, including elastic, buttons, zippers, and similar items. Approximately 16 firms produce woven fabric, while 15 companies produce knit fabric. Of the country’s textile producing firms, nine are vertically integrated, manufacturing either yarn through fabric, yarn through finished apparel, or yarn through household textiles. Cotton, wool, mohair, manmade fibers, and natural fibers are used in the domestic textile industry.	Competitive advantages: • large domestic apparel industry; • developed infrastructure (transport, power, water); • favorable and stable business environment; • large and developed textile industry; Competitive disadvantages: • high labor costs; • inflexible labor market; • lack of skilled labor in the industry • lack of management, marketing, and technical skills; • lack of investment; • long lead times from order to delivery; • highly volatile exchange rate
Swaziland	Swaziland has one integrated textile producer that dyes, spins, and knits cotton fabric (including organic), and then sews the fabric into apparel for export. The firm produces yarn for internal consumption and for export to the region and the EU. Swaziland has an internationally branded zipper producer that supplies local and regional apparel manufacturers.	Competitive advantages: • export-oriented apparel industry; • government incentives for foreign direct investment in the textile and apparel industry; • reliable electricity supply; Competitive disadvantages: • small domestic apparel market; • limited amount of local raw materials; • labor unrest; • high HIV/AIDS prevalence rates
Tanzania	The Tanzanian textile sector consists of one independent spinning mill and several integrated firms. The industry spins mostly cotton yarns for both knit and woven fabric. A few fabric mills also blend cotton with polyester or other synthetic fibers; however, all synthetic fibers must be imported. Tanzanian textile mills sell these textiles regionally, or minimally process and print fabric to be sold locally as final products.	Competitive advantages: • availability of good-quality domestic cotton; • history of cotton yarn exports to the EU; • stable political and economic environment; Competitive disadvantages: • lack of a developed apparel industry; • unreliable and costly electricity; • port delays and congestion; • lack of skilled labor; • lack of market knowledge; • low labor productivity
Zambia	The Zambian textile sector consists of an estimated four knitting/weaving firms and four vertically integrated firms that spin their own yarn for use in finished textile and apparel production. Zambia’s textile sector produces primarily 100 percent cotton yarn, along with small quantities of manmade-fiber yarn, including poly/cotton and acrylic yarn. Most of the yarn produced in Zambia is exported, but a small share is used domestically in the production of woven fabric used to manufacture niche apparel articles such as uniforms and mining work wear, primarily for the local or regional market.	Competitive advantages: • domestic availability of high-quality cotton; • open trade regime; Competitive disadvantages: • small domestic apparel market; • insufficient access to affordable credit; • outdated machinery and equipment; • lack of skilled labor; • low labor productivity; • high transportation costs and time • unreliable electricity supply

Table 3.

Summary of selected SSA textile and apparel input producers.

Changes in the volume by country are illustrated graphically on the Figure A1 on Appendix 1.

The African Value Chain is highly fragmented and is illustrated by Figure 10 below.

Figure 10.

The African textile value chain.

The 4 leading countries on imports of apparels from SSA to USA under AGOA are Kenya, Lesotho, Madagascar and Ethiopia. Opportunities for development of the textile-cotton industry in SSA depend on the following critical success factors. African countries face 5 key opportunities to develop the cotton-textile sector and these are:

1. Restructuring (Shift of industry from China to other developing and LDC’s; consolidation and upgradation; economic imperative, etc.)

2. Endowment (Availability of abundant raw material; Africa’s labour pool; Land; Water, etc.)

3. Market access (Preferential market access to the EU and US; RTA’s/AfCFTA; etc.)

4. Global initiatives (Belt & Road Initiative; other international projects, etc.)

5. Sustainability (Guidance from the 17 SDGs).

These opportunities require policy actions in the following areas:

1. Enabling environment

2. Market access

3. Raw material

4. FDI

5. Capacity building

The textile industries of Ghana, Nigeria, Uganda, and other nations have been destroyed by cheap Chinese imports. According to the Industrial and Commercial Workers Union in Ghana, only four out of thirty textile enterprises are still active (ICU). The organization claims that the country used to produce yarn for garments marketed domestically and in Sub-Saharan Africa, but that this is no longer the case.

8. Conclusion and recommendations on the cotton and textile industry in SSA

Apparel manufacturing is particularly labour-intensive, with minimal start-up costs and readily transferable technologies. As a result, several nations with low labour costs, particularly in South and East Asia, have gained significant market share in the recent four decades.

Main policy recommendations for LIC governments, industry associations and clothing firms can be summarized as follows:

1. Improve productivity, skills, and capabilities within fi rms and develop from cutmake-trim (CMT) to full package suppliers.

2. Increase backward linkages and reduce lead times.

3. Improve physical and bureaucratic infrastructure.

4. Improve labour and environmental compliance.

5. Diversify end markets to fast-growing emerging markets.

6. Increase regional integration.

7. Build locally embedded clothing industries.

8.1 Big data analytics framework model for oilseeds and textile production in SSA

From the concept of a Big Data strategy to the technical tools and capabilities that a company should have, there’s a lot to consider. The following are the key advantages of using a Big Data framework:

1. The Big Data Framework provides a framework for businesses looking to get started with Big Data or improve their Big Data capabilities.

2. The Big Data Framework encompasses all aspects of an organization’s structure that must be considered in a Big Data environment.

3. The Big Data Framework is not tied to any particular vendor. You can expand the data storage by adding more nodes.

The Big Data Framework’s Structure.

When establishing a Big Data organization, organizations should consider the Big Data framework, which is a structured approach that comprises of six basic capabilities. The following is a diagram of the Big Data Framework shown on Figure 11. The Hadoop Architecture layout is shown on Figure 12.

Figure 11.

Big data analytics framework model for oilseeds and textile production.

Figure 12.

Hadoop’s architecture.

When establishing a Big Data organization, organizations should consider the Big Data framework, which is a structured approach that comprises of six basic capabilities. It’s a people business when it comes to big data. Even with the world’s most modern computers and processors, businesses will fail if they lack the necessary knowledge and skills. As a result, the Big Data Framework strives to broaden the expertise of everybody interested in Big Data. The modular method, as well as the supporting certification scheme, attempts to create Big Data knowledge in a similar organized manner. The Big Data framework is a comprehensive approach to Big Data. It examines the numerous elements that businesses should consider when establishing a Big Data company. Every component of the framework is as important, and organizations can only progress if they give all components of the Big Data framework similar attention and effort.

9. Conclusion

The Hadoop platform was created as a framework for big data analytics. The open-source Hadoop platform offers the analytical tools and computing capacity needed to handle such massive data volumes. The Hadoop Distributed File System (HDFS) and the MapReduce parallel processing engine are the two primary parts of Apache Hadoop. Apache Hadoop has been successfully established as an open source option for distributed systems in the fields of Big Data, cluster, and cloud computing. Scalability, availability, and fault tolerance to a great degree are promised by the master/slave design. By simply adding existing hardware, it is possible to obtain cost-effectively extra memory, increased I/O capacity, and improved performance. A technology called Map-Reduce allows for the concurrent processing of sizable data sets across many nodes in sizable clusters. Map-Reduce at the level of “Distributed data processing” coupled with the database “HBase” can be taken into consideration since the processing and management of data are two things that are naturally in direct connection.

Because they are self-pollinated crops and farmers can keep and recycle grain from past harvests, the competitiveness of oil crop or legume seed markets is limited by the poor rate of return on investments in breeding, seed production, processing, and marketing. One way to do this is to persuade commercial seed companies to invest in seed production of publicly developed varieties, and to work with them and other stakeholders to improve coordination along the value chain in order to provide farmers with the necessary incentives to invest in improved seed and other complementary inputs to increase productivity and improve quality. The major textile firms produce a variety of products, including polyester staple fiber and filament, yarn, greige cloth, and wax prints.

Acknowledgments

Great appreciation is expressed to the University of Zimbabwe Business School Masters students who studied the Applied Business Informatics module taught by the author from April to July, 2022 and who participated in the Focus Group discussions.

Figure A1.

Changes in the volume of production by country.