Innovations in science and technology have allowed our Kenyan telecommunications industry to expand services, reduce prices, and grow to meet the ever increasing demands from other sectors—sectors that increasingly rely on telecommunicatio


ns to help them work efficiently while reducing their energy consumption. The current worldwide focus to reduce energy consumption and ecological impacts is adding another dimension to the business cases for new network technology, implementation of energy-saving systems, and installation of green alternative power solutions. All these requires a plan based on network architectures, service offerings, customer base, and geographic location. This is a challenging, multi-dimensional planning process that I personally am very eager to take up as an engineer and give solutions to help my fellow Kenyans and the entire country at large. Planning an efficient, eco-sustainable telecommunication


s network while meeting corporate financial and business goals is a challenging problem. While networks are naturally evolving more so in east Africa to become more efficient, this am afraid will take years to occur(note as a country, we have made bigger stride


s on the evolution of our networks as per CAK 2017 report). With the current high energy costs and government and consumer interest in greener technology, proactively reducing energy consumption and adding alternative, eco-friendly power sources often result in a better business plan, both short term and long term. Deciding on specific energy reduction initiatives and determining the viability of alternative power options that meet network operator business goals require a complex, multidimensional analysis that I should clearly state forthright that Safaricom PLC has already identified these fields and has started off tapping into giving solutions aligned together with the Paris agreement as well as the 2015 SDGs.



My personal goal: I want and will come up with a solution(s) to help Kenyan network operators’ work through the maze of conflicting requirements and options to identify eco-friendly solutions that meet the financial and business goals while reducing the ecological impacts.


Written and complied by: Samwel Kariuki


It is a worldwide goal to reduce energy consumption and CO2 emissions. The EU has targeted a reduction of 20% for year 2020 and just the other week we saw an MoU between Safaricom and UN signed championing for SDGs set.. A part of this energy reduction scheme concerns the    telecommunication industry and ICT that participates in a direct, indirect and systematic way.  Characteristic  examples which are yet to be in full use or are at nascent stages in our country  are  green networks,   smart   buildings, 20170820_152219.jpg  smart   grids,   Intelligent Transportation Systems (ITS), energy efficient electronics (OLEDS, photonics, nanotechnology) and the application of  embedded  systems  towards  low  carbon  and  energy efficient technologies .

Telecommunication networks constitute a major sector of ICT and they undergo a tremendous growth. Capacity issues and delivery of complex real time services are some of the main concerns that yield high power consumption patterns. In our increasingly competitive mobile telecommunication sector, operators are turning to emerging markets for their next step growth situation that increases the number of subscribers and required base station equipment-case examples include safaricom now on 4G+ while Telkom is rolling our 4G across its country’s network footprint. This creates the need for equipment installation to areas where off grid renewable energy solutions are required and energy efficient networks are important e.g. Northern parts of Kenya. In addition, the increase of fuel and electricity costs bounds the OPEX of the system.

Telecommunication networks and broadband access are proved to consume a huge amount of energy for data delivery.  In general, the telecommunication   sector accounts for approximately 4% of the global electricity consumption (I researched widely from ITU web links).  I personally believe that  reduction  of  CO2   emissions  can  be accomplished  by  focusing  on  innovative telecommunication services like online taxation, video conference,   online   billing   that   can   enable   a   green economy. The goal is to deploy telecommunication networks enabling power efficiency, yielding a small ratio of required Watts per Gbps and Watts per user. Green initiatives have already been commenced by different operators around developed countries.

This summarized word press discusses and proposes various energy efficient techniques for the green operation of telecommunication networks. Cellular networks that suffer most of the power waste nowadays are what I will highlight briefly. It is observed that almost 50% (including the operation of servers) is due to the operation of telecommunication networks. These can be mobile networks, WLANs, LANs and fixed line networks. As  far  as  the  overall network  performance  is  concerned  the  energy consumption is higher at the access part of the network and the operation of data centers that provides computations, storage, applications and data transfer in a network. On the other hand, backbone and aggregation networks present lower energy demands. This makes clear that an energy efficient architecture should focus on intelligent and efficient access techniques and efficient operation and data manipulation by data centers. The main functionalities of a network can be summarized as the process of regeneration, transportation, storage, routing, switching and processing of data. The power consumption patterns of these processes can be observed that the largest part of energy is consumed for routing/switching, regeneration and processing of data. Both communication protocols and electronic devices are responsible for this consumption and this imposes challenges for more sophisticated transport techniques, thermal removal from switches or the servers and less redundant data transfers.

For mobile networks, a crucial factor affecting network power consumption is the site operation that incorporates base station equipments. . It is obvious that the greatest portion of energy is consumed for cooling of equipments and base station operation. Monitor operation and lighting requires the minimum of energy whereas for the backhaul energy consumption the picture is not clear and depends on the type of connections of the backhaul network (fiber or cable).  Within the base stations, high power demands are due to feeders (transmission of radio waves), the RF conversion units and power amplifiers, signal processing units and various electronic   equipments   such   as   air   conditioners   and auxiliary equipments.

The power consumption within a base station exhibits important similarities with data centers. The available power from the electricity grid, the battery backup unit or the renewable energy (RES) enters the base station and is divided into an in-series path and an in-parallel path. Non- critical equipments support the operation of the IT equipments that are divided into radio units and baseband units. The most energy consuming devices of base stations are the cooling infrastructure, power amplifiers, RF feeders and the AC/DC and DC/DC conversion units. Depending on the number of sectors, nSC, and the antenna number, nTX, of the base station, the total power consumption is computed as follows;



In the above formula an additional factor models the power consumption due to RF links of the base station. For macrocell and microcell base stations, empirical formula can describe the relationship between the power delivered to the antenna relative to the consumed power of the base station [13]. For macrocell stations the power consumption is almost independent of the input load (traffic) whereas for microcells, power consumption is highly dependent on the input load.

Making a network to operate in a green manner is a complex task. Sometimes, optimizing energy consumption in one part of the network can increase power consumption and degrade the performance of another part of the network. In general, total network optimization is better than the sum of optimizations of individual parts. A network to work in an energy efficient way is not only a matter of environmental protection through signing of memorandums but also a crucial factor for the deployment of future networks to off grid areas that rely on Renewable Energy Sources (RES) or personal and sensor networks that rely on battery power supply. Minimizing power consumption has also a great effect on the cost of operation of a network and this makes it more affordable to the user. Network energy efficiency can be considered as a very complex task since there is no clear solution to the problem. The sectors of the network that require the greatest attention are the electronic equipments of both end user and the access network, thermal removal processes, efficient network planning and base station design.


Compiled and written by: Samwel Kariuki

Technology is changing rapidly for wireless, significantly changing the power requirements of the 6000+ base stations within our Kenyan Telcos infrastructure. These improvements increase the viability of using Eco-friendly power and our Telcos have img-20161025-wa0004.jpgalready seen this trend of IoT and are engaged in efforts to stop the trend of rising telecom energy demands. With so many options for reducing their Eco-footprint, and considering the challenge of implementing changes while remaining profitable, planning a sensible, ecologically friendly path forward is often a formidable task. It is for this reason that I chose to take an opportunity to write to the power departments in our communication institutions which I have gracefully worked with for close to 3 years indirectly as an engineer assigned to do electrical and computational works for them.

The4G+ as an already laid out plan by one of the major Telcos within our country serves as an example which is a really good move that comes with growth of bandwidth demand which can easily cause Safaricom network  energy  consumption to  rise  in  step  with  the growth. The resulting increase in electricity costs leads to reduced margins at a time when competition is also driving prices down-the relauch of Telkom Kenya a few days ago marks a threat in the same regard. Having worked closely with a number of power departments amongst the Telcos we have, I have seen and learnt two options used when planning to reduce power consumption.  First, there are new network architectures that are inherently more energy-efficient and which can simultaneously provide the flexibility to support continued increases in demand. Second, choices in network equipment, options, and support equipment for new or existing infrastructure have also had a tremendous impact on the amount of power consumed. Both options are quite viable and should be part of any power reduction plan even as we leap into the digital disruptive era in the coming years.

Am grateful to have worked indirectly with the engineers at both power and optimization departments and have been able to tap a lot of skills in my area of expertise and personal growth as well. I look forward for an opportunity to present my ideas (a combo mixture of Artificial intelligence, big data analytics and IoT) as well as deliberate further on how best can power can be planned and supported to attain the ultimate goal in energy efficiency. Am also grateful to Parastatals that deal directly and indirectly with power and energy distribution for the nifty work they are putting across to solve the trilemma of cost, reliability and quality of power being used in our republic.

Below is a recap article of the latest bell lab power technical journal 2017 edition that I saw it prudent to share as well with other engineers and stakeholders in power & energy sector alike whom I revere and hold atmost respect for the trainings and lessons I have gained from them.


Methodology for Planning Energy-Reducing

The methodology for planning network changes to reduce energy usage consists of three cascading steps:

  • Energy consumption hierarchy. Identification of the network elements that consume power and their location in the network.
  • Energy-saving chain. Identification of network element dependencies upon each other’s  power  dissipation (e.g.,  larger  air  conditioning units  having  higher energy   consumption are  necessary if inefficient power   rectifiers   are   installed  because  of  the energy  they  waste  through heat  radiation).  This allows network operators to target the most effective points for energy reduction by applying energy-saving initiatives.
  • Energy-saving initiatives or options. Determination of specific choices or actions that  can be taken to reduce energy  consumption for one or more  net- work elements (e.g., replacing low-efficiency rectifiers with high-efficiency rectifiers, which requires capital  and  installation expense,  but these  expenses may be offset in 12 to 15 months based  on  today’s  high  energy  costs).  Sets of initiatives are often deployed simultaneously due to typically lower installation costs as compared to deploying the initiatives one at a time


As we continue improving our communication systems across the country and beyond, lets research and read widely for the upcoming 4th industrial revolution which in my own view will be sparked and born in China & embraced fully here in Africa and hopefully in our dear motherland Kenya.




“A powered nation is a growing nation”~Samwel Kariuki

Data Mining is the process of sifting through stores of data to extract previously unknown, valid patterns and relationships that provide useful information. Once these patterns are found they can further be used to make certain decisions for development of businesses. For decades major components of data mining img_2337.pngtechnology have been under development in research areas such as statistics, artificial intelligence and machine learning. But in recent times we are yet to see here in Kenya the maturity of these techniques coupled with high performance relational database engines and broad data integration efforts make these technologies more effective for current data warehouse environments within our telcos.

Telecommunication data pose several interesting issues for data mining more so in our now-growing digitalization exodus that a lot of Kenyans if not most are experiencing. The first concerns scale, since telecommunication databases may contain billions of records and are amongst the largest in the world. A second issue is that the raw data is often not suitable for data mining. For example, both call detail and network data are time-series data that represent individual events. Before this data can be effectively mined, useful “summary” features must be identified and then the data must be summarized using these features. Because many data mining applications our telecommunications industry involve predicting very rare events, such as the failure of a network element or an instance of telephone fraud, rarity is another issue that must be dealt with. The fourth and final data mining issue concerns real-time performance: many data mining applications, such as fraud detection, require that any learned model/rules be applied in real-time.

We can distinguish three main steps of describing data mining problems:

  1. Problem formulation in the domain terms. This is usually textual description of the business requirements that have to be fulfilled by data mining.
  2. The transformation of business requirements into a class of data mining problems like classification, prediction, associations etc. It is a bridge between business description and detailed model specification.
  3. The detailed model specification. This is a model specification that is used by data mining modeler for a specific software tools.

It can be observed that one of the main areas of applications of data mining on business level is a support for various tasks of the marketing departments. The data mining becomes a key part of analytical subsystem of customer relationship management systems. On business level of analysis there are many similarities to other industries. The following main problems for example in marketing and sales departments of telecommunication operators can be distinguished: – customer segmentation and profiling, – churn prediction, – cross selling and up-selling, – live-time value, – fraud detection, – identifying the trends in customer behavior. On product or service level there is a focus on analysis of incomes, quality of the service, grade of the service and others. There are formal agreements called service level agreements (SLA) between providers of the service and the customers. Service level management (SLM) is becoming the prevailing business model for delivering a products and services. Such approaches need advanced computerized tools. On the level of infrastructure and network analysis we can distinguish the following problems: – network planning, – IT resources planning, –: fraud detection, marketing/customer profiling and network fault isolation.

One central issue is that telecommunication data is often not in a form—or at a level—suitable for data mining. Other data mining issues that are or/and will affect our Kenyan telcos include the large scale of telecommunication data sets, the need to identify very rare events (e.g., fraud and equipment failures) and the need to operate in realtime (e.g., fraud detection). Data mining applications must always consider privacy issues. This is especially true in the telecommunications industry, since telecommunication companies maintain highly private information, such as whom each customer calls. Personally am elated that most telecommunication companies in our country utilize this information conscientiously and consequently privacy concerns have thus far been minimized. A more significant issue in the telecommunications industry relates to specific legal restrictions (provided by CAK) on how data may be used. In the United States, the information that a telecommunications company acquires about their subscribers is referred to as Customer Proprietary Network Information (CPNI) and there are specific restrictions on how this data may be used. It’s in Kenyans interest that the communication Authority of Kenya (CAK) generally sticks to prohibiting the use of that information without customer permission, even for the purpose of marketing the customers other services. In the case of customers who switch to other service providers, the original service provider is prohibited from using the information to try to get the customer back (e.g., by only targeting profitable customers). Furthermore, all our CSP (communication service providers) & ISP (internet service providers) should be prohibited from using data from one type of service (e.g., wireless) in order to sell another service (e.g.. landline services). Thus, the use of data mining is restricted in that there are many instances in which useful knowledge extracted by the data mining process cannot be legally exploited. Much of the rationale for these prohibitions relates to competition. For example, if one of our telcos can leverage the data associated with one service to sell another service, then a number of Kenyan companies (both CSPs & ISPs) that provide fewer services would be at a competitive disadvantage.

Kazi kwenyu CAK as we anticipate for a colossal unprecedented digital migration that will in hand come bearing babies like data mining, Artificial intelligence, cognitive cloud computing and all sorts of intelligent smart technologies with them. As for our Kenyan telcos, I strongly believe having worked in this industry for more than 3+ years that platforms and foundations (software and hardware) are already set in place in anticipation for the coming 21st exodus.


Complied by: Samwel Kariuki

Dated: 23rd May 2017

There has been recent progress in the analysis of call-center data.   Call-by-call  data  from a small number  of sites  have  been  obtained  and  analyzed,  and  these  limited  results  have  pr…

Source: Future Work in Data Analysis and Forecasting within our Kenyan Telcos and Africa at large.

There has been recent progress in the analysis of call-center data.   Call-by-call  data  from a small number  of sites  have  been  obtained  and  analyzed,  and  these  limited  results  have  proven  to  be fascinating.    In  some cases,  sucimg_3067h  as the  characterization of the  arrival  process  and  of the  delay of arriving  calls to the  system,  conventional assumptions and models of system  performance  have been upheld.  In others, such as the characterization of the service-time distribution and of customer patience, the data have revealed fundamental, new views of the nature of the service process.  Of course, these limited studies are only the beginning, and the effort to collect and analyze call-center data can and should be expanded in every dimension in Kenya and Africa at large.

Perhaps the most pressing practical need is for improvements in the forecasting of arrival rates. For highly utilized call centers, more accurate, distributional forecasts are essential.   While  there exists  some research  that develops  methods  for estimating and  predicting  arrival  rates, I strongly believe there  is surely room for additional improvement to be made both here at home and the entire continent.  However, further development of models for estimation and prediction will depend, in part, on access to richer data sets.  Some of  us believe that much of the randomness of Poisson arrival rates may be explained by covariates that are not captured in currently available data.

Procedures for predicting waiting-times are also worth pursuing.  Field-based studies that characterize the performance of different statistics and methods would also be of value.  More broadly, there is need for the development of a wider range of descriptive models.  While a characterization of arrival  rates,  abandonment  from queue,  and  service times  are essential  for the  management  of call centers,  they constitute only a part  of the complete picture  of what goes on. For example, there exist (self ) service times  and abandonment (commonly  called “opt-out”) behavior  that arise from customer  use of IVRs.  Neither of these phenomena is likely to be the same as its CSR analogue. Similarly,  sojourn  times  and  abandonment from  web-based  services  have  not  been  examined  in multi-media centers.


Parallel, descriptive studies are also needed to validate or refute the robustness of initial findings. For example, lognormal service times have been reported in two call centers, both of which are part of retail financial services companies.   Perhaps the service-time distributions of catalogue retailers or help-desk operations have different characteristics.

Similarly, one would like to test some finding that the waiting-time messages customers hear while tele-queueing promote, rather than discourage, abandonment.

It would also be interesting to put work on abandonment (Palm, Roberts, Kort, Mandelbaum with Sakov and Zeltyn)  in perspective.  These studies provide empirical and exploratory models for (im)patience on the phone in Sweden in the 40’s, France  in the late 70’s, the U.S. in the early 80’s, Israel in the late 90’s and  now Africa(Kenya in particular under this research) in the early millennium. A systematic comparison of patience across countries, for current phone services, should be a worthy, interesting undertaking.

There is the opportunity to further develop and extend the scope of explanatory models.  Indeed, given  the  high  levels of system  utilization in  the  QED Quality  and Efficiency Driven (operational)  regime,  a  small  percentage  error  in  the forecast  of the  offered load can lead to significant,  unanticipated changes  in system  performance. In particular, the state of the art in forecasting call volumes is still rudimentary. Similarly, the fact that service times are log normally distributed enables the use of standard parametric techniques to understand the effect of covariates on the (normally distributed) natural log of service times.

In well-run QED  call centers,  only a small fraction  of the  customers  abandon (around 1-3%), hence about  97% of the  (millions  of ) observations  are  censored.   Based on such figures, one can hardly expect any reasonable estimate of the whole patience distribution, non-para-metrically at least.  Fortunately, however, theoretical analysis suggests that only the behavior of impatience near the origin is of relevance, and this is observable and analyzable.

Indeed,  call-center  data  are  challenging  the  state-of-the-art of statistics, and  new statistical techniques  seem  to  be  needed  to  support their  analysis.    Two  examples  are  the  accurate   non- parametric estimation of hazard  rates,  with  corresponding  confidence intervals,  and  the  survival analysis  of tens  of thousands, or  even  millions,  of observations, possibly  correlated   and  highly censored.

Last but certainly not least, a broader goal should be, in fact, the analysis of integrated operational, marketing, human resources, and psychological data.  That is, the analysis of these integrated data is essential if one is to understand and quantify the role of operational service quality as a driver for business success.

A prerequisite for understanding the financial effects of operational decisions is the ability to analyze an integrated data set that includes operational (ACD) automatic call distributor and marketing / business (customer information systems) data.   With this information, one can attempt to tease out the longer-term, financial effects of operational policies.

My experience  has  been  that both  types of data  are  very  difficult to  access,  however. One reason for this is technical. Only recently  have  the  manufacturers of telephone  equipment given customers  something  of an “off the  shelf ” ability  to capture, store,  and  retrieve  detailed,  call-by- call data.    Similarly,  the  integration of these  operational data  with  the  business  data  captured in customer  information systems  is only now becoming  widely available.   Another reason stems from confidentiality concerns; most of our Kenyan companies are rightly wary of releasing customer information.  Once managers recognize the great untapped value of these data, i believe they will employ mechanisms for preserving confidentiality in order to reap the benefit.

Ultimately, i envision a data-repository that is continuously fed by many call centers of varying types.  The collected data would be continuously and automatically analyzed, from both operations and marketing perspectives.  Then the data  would be both archived and fed back to the originating call centers,  who would use it (through visualization tools) to support ongoing operations, as well as tactical  and strategic goals.

Little imagination is required for appreciating the value of such a data-base.  As a start, its developer could become a benchmark that sets industry standards, as far as customer-service quality and call-center efficiency are concerned.  As already mentioned, such a data-base would enable the identification of success-drivers of call-center business transaction.



Researched & Compiled: Samwel Kariuki                                                                                                                                                                                                                                                                          Date: 12th Feb 2017


5G, WiFi, GPRS, NB-IoT, LTE-M & LTE Categories 1 & 0, SigFox, Bluetooth, LoRa, Weightless-N & Weightless-P, ZigBee, EC-GSM, Ingenu, Z-Wave, Nwave, various satellite standards, optical…

Source: Future Prospects in IoT / M2M for our Kenyan Telcos


5G, WiFi, GPRS, NB-IoT, LTE-M & LTE Categories 1 & 0, SigFox, Bluetooth, LoRa, Weightless-N & Weightless-P, ZigBee, EC-GSM, Ingenu, Z-Wave, Nwave, various satellite standards, optical/laser connections and more….. The list of current or proposed wireless network technologies for the “Internet of Things” seems to be growing longer by the day. Some are long-range, some short. Some high power/bandwidth, some low. Some are standardized, some proprietary. And while most devices will have some form of wireless connection, there are certain categories that will use fibre or other fixed-network IMG_2340.pnginterfaces.

There is no “one-size fits all”, although most Kenyans hope that 5G will ultimately become an “umbrella” for many of them, just like the current 5x speeds offered today by Safaricom under 4G(#DoingItTheSafaricom4G) . But our Kenyan telcos, especially mobile operators, need to consider which they will support in the shorter-term horizon, and for which M2M/IoT use-cases. That universe is itself expanding too, with new IoT products and systems being conceived daily, spanning everything from hobbyists’ drones to industrial robots(as of the case in Mitsubishi IIoT recent innovations done at Centurion Systems, westlands).. All require some sort of connectivity, but the range of costs, data capabilities and robustness varies hugely.

Two over-riding question themes emerge:

  • What are the business cases for deploying IoT-centric networks – and are they dependent on offering higher-level management or vertical solutions as well? Is offering connectivity to Kenyans– even at very low prices/margins – essential for telcos to ensure relevance and differentiate against IoT market participants?
  • What are the longer-term strategic issues around telcos supporting and deploying proprietary or non-3GPP networking technologies? Is the diversity a sensible way to address short-term IoT opportunities, or does it risk further undermining the future primacy of telco-centric standards and business models? Either way our telcos need to decide how much energy they wish to expend, before they embrace the inevitability of alternative competing networks in this space.

Personally I’ve outlined three strategic areas of M2M business model innovation for our Kenyan telcos:

  • Improve existing M2M operations: Dedicated M2M business units structured around priority verticals with dedicated resources. Such units allow telcos to tailor their business approach and avoid being constrained by traditional strategies that are better suited to mobile handset offerings. A good working case scenario is what Safaricom is currently doing with their VAS services.
  • Move into new areas of M2M: Expansion along the value chain through both acquisitions and partnerships, and the formation of M2M operator ‘alliances.’ Here in kenya,I should give Microsoft and Salesforce from America a thumbs up for the nifty partnership they are doing to realize this very important point.
  • Explore the Internet of Things: Very few of our telcos have been active in the connected home e.g. THE BIG BOX from Safaricom. However, outsiders are raising the connected home (and IoT) opportunity stakes: Companies from Thailand, Canada and china are flooding our market with android TV boxes and such more connected homes.


  • 2015 discussion of IoT connectivity has been dominated by futuristic visions of 5G, or faster-than-expected deployment of LPWANs (low-power wide-area networks), especially based on new platforms such as SigFox or LoRa Alliance.
  • As I write this article, already in Kenya there is a comparatively slow emergence of dedicated individual connections for consumer IoT devices such as watches / wearables. With the exception of connected cars, most mainstream products connect via local “capillary” networks (e.g. Bluetooth and WiFi) to smartphones or home gateways acting as hubs, or a variety of corporate network platforms. The arrival of embedded SIMs (Safaricom being on the fore front in deploying the first bunch countrywide) might eventually lead to more individually-connected devices, but this has not materialised in volume yet.
  • Growing impatience among some in the telecom industry with the pace of standardization for some IoT-centric developments. A number of operators have looked outside the traditional cellular industry suppliers and technologies, eager to capitalize on short-term growth especially in LPWAN and in-building local connectivity. In response, vendors including Huawei, Ericsson and Qualcomm have stepped up their pace in our country, although fully-standardized solutions are still some way off.

This highlights the IoT quantification dilemma – most if not all Kenyan telcos are focusing on the big numbers, many of which are simple spreadsheet extrapolations, made without much consideration of the individual use-cases. And the larger the headline number, the less-likely the individual end-points will be directly addressed by telcos.


Compiled by: Samwel kariuki

18th Nov 2016



Posted: October 13, 2016 in Uncategorized

Around 5.5 million new things will get connected to the Internet of Things (IoT) every day during 2016. Research also predicts that 6.4 billion connected things will be in use worldwide in 2016, up…



Posted: October 13, 2016 in Uncategorized

Around 5.5 million new things will get connected to the Internet of Things (IoT) every day during 2016. Research also predicts that 6.4 billion connected things will be in use worldwide in 2016, up 30% from 2015.img_3068

Many of the items currently associated with IoT — things like connected thermostats, fitness trackers, connected cars and automated home appliances — are primarily for consumers. However, in Kenya, IoT technology could have an even bigger impact on enterprises than on individuals. This is for obvious reasons of financial strains folks are undergoing through unlike enterprises which keep making profits thus can invest on such emerging technologies.

For enterprise IT departments, IoT will offer a number of challenges. They will need to deploy and maintain the sensors and IoT gateways that will make up the Internet of Things. They will need to develop and support a variety of new IoT applications for smart manufacturing, asset tracking, smart offices, and other purposes. IT teams also will have to make sure they have the necessary networks and systems to handle the huge increase in data, as well as the tools to process and secure all that information.

Because IoT is still in its infancy in Kenya, most organizations are still trying to figure out how it will affect them. IoT has gained mainstream awareness, yet organizations are still struggling with how to deal with the complexities of the vendor ecosystem in terms of developing and deploying connected products and services.

Connected manufacturing

IoT is expected to make such a big impact on manufacturing processes that many vendors and industry watchers are likening it to another industrial revolution. A 2015 survey found that 40% of manufacturers believe that smart manufacturing technology is ready for use and now is the right time to invest……moreso in Kenya & SA where technologies are embraced every day and used optimally. The new technology will allow companies to have visibility into every step of the manufacturing process and enable them to track each individual item as it is produced. In addition, automated systems will be able to tailor production to meet demand and to order supplies in real time as they are needed.

But before smart manufacturing technology can hit the factory floor, enterprises will need to update their other systems to track and analyze the new data.

Asset tracking

Another key enterprise application of IoT technology will be asset tracking. Startups and established companies alike are offering sensors for tracking just about anything a company could own — fleet vehicles, retail products, IT systems, industrial equipment, supplies and more. Of course, companies will then also need cloud-based systems to aggregate and report on the data that comes in. Vendors like BlackBerry are rolling out end-to-end IoT asset-tracking products in hopes of taking a leading role in this new market.

Advertising and marketing

For Kenyan marketers and retailers e.g.  Nakumatt, Naivas, Tuskys and major Chain restaurants, IoT sensors embedded in the products that they sell could supply a wealth of data about how their products are used. For example, a smart blender could provide information about what time of day it is used, how frequently it runs, how long it is on, and when it breaks down. They can then use that information to improve customer service, enhance their products, and create more effective advertising.

In addition, some retailers are experimenting with in-store beacons that allow them to track customer movement in stores and push advertising and other content to smartphone users.


Smart offices

A lot of IoT talk has focused on smart homes, but for enterprises, IoT also brings up the possibility of smart offices. Companies like Enlightened offer sensors that can track when people are in certain parts of buildings. That information can be used to control lighting and temperature controls, or it could be sent to enterprise analytics systems to help companies make decisions about how much square footage they need, what size conference rooms get used the most and other real-estate issues. Similar technology is also helping retailers learn more about how customers and staff interact in stores



For consumers, one of the most exciting possibilities of the Internet of Things is connected and even self-driving cars, like the Google Self-Driving Car Project. Eventually, autonomous vehicles like these could become a major part of enterprise fleets and play a much larger role in transporting goods from one place to another.

But long before fully autonomous vehicles become commonplace, enterprises will be connecting their fleets of vehicles to the Internet in order to better track the supply and delivery of goods. Logistics companies, delivery services, trucking companies, retailers, wholesalers, distributers, utilities and many other kinds of enterprises are beginning to collect and track fleet data, analyzing it for ways to make their operations more economical and efficient.


Smart electricity use

For many organizations in Kenya, particularly manufacturers and companies with large data centers, electricity is a major expense. Perhaps it shouldn’t be a surprise then, that one 2015 report found that “the lowly energy meter is becoming a leading device in the transition to the Internet of Things.” It estimates that there were 454 million smart meters installed in 2015 and that there will be twice as many in use by 2020.

In addition to smart meters, businesses are employing a variety of other sensors designed to help them use energy more efficiently. For example, in data centers, temperature sensors can allow them to direct air conditioning towards the areas where it is needed or to distribute virtualized workloads to physical servers dispersed throughout a facility in or order to reduce or eliminate the need for cooling.


Network expansion

As enterprises adopt more IoT applications, they will have large quantities of sensors and IoT gateway devices connected to their networks, and the amount of data flowing through corporate networks will increase dramatically. And that means organizations will need to beef up their networking hardware and available bandwidth. For many enterprises, widespread adoption of IoT will require the purchase of additional networking equipment and perhaps the hiring of additional network administrators in order to manage and maintain the new equipment.


More data and analytics

Not only will companies need to enhance their IT networks in order to deal with the deluge of IoT data, they’ll also need to upgrade their storage systems and improve their big-data analytics capabilities. For many, this will mean investing in next-generation analytics solutions that feature cognitive computing or machine learning capabilities.

IoT security

Personally as i have noted, security remains the big question mark for IoT deployments. Organizations are still figuring out how to secure all the devices, sensors and data that will make up the Internet of Things. Based on my analyst estimates of the number of connect things and the size of the IoT security market in Kenya at current times, it will cost about 100 KSH per thing to secure the Internet of Things. As organizations look to deploy thousands, or even millions, of IoT devices and sensors, those security costs could ramp up quickly. But the potential cost to companies that fail to secure their IoT deployments could be even greater.


Compiled & written by: Samwel Kariuki