DEV Community: Aleksei Badianov

Driving Sustainability through Route Optimisation and Delivery Management

Aleksei Badianov — Fri, 12 May 2023 16:47:22 +0000

Introduction

According to a report by Statista, the UK e-commerce market has been steadily growing, with sales amounting to £161 billion in 2022 and projected to reach £161 billion by 2025. Home delivery has played a significant role in this growth, with sales giants like Amazon, Ocado, and eBay leading the way. In fact, in 2020, the global online grocery market size was valued at $190.2 billion, a significant increase from $98.4 billion in 2019, according to ResearchAndMarkets.com.

The COVID-19 pandemic further accelerated the growth of home delivery, as households worldwide were forced to limit their physical movements. In the United States, for example, online spending for home delivery increased by 30% in 2020, according to a report by the National Retail Federation. This surge in demand contributed to the growth of the last-mile delivery sector.

The impact of the pandemic also caused many businesses to switch from B2B to B2C delivery models to survive. According to a survey by Accenture, 60% of businesses reported that the pandemic has accelerated their plans to invest in direct-to-consumer channels. This shift has further fueled the growth of last-mile delivery, as manufacturers and wholesalers now need to deliver directly to consumers.

Route optimisation

One of the primary bottlenecks in new delivery models is delivery planning. A good delivery process always starts with a delivery schedule or plan. With B2B organisations, those plans were polished within years to make them logical, simple to implement, and efficient. Fixed routes were well balanced to consider average volumes, business customer ranking, busy hours individual for each delivery location, and could be reused for many years.

With a change to B2C model delivery, planning became a daily routine as the volume of orders fluctuated, and the same household did not repeat the same order every day. Consequently, the same mechanism that historically worked for B2B planning was not helpful in the B2C world.

The complexity of route planning increases exponentially with the number of stops in a route. For instance, a route with four stops has only 24 potential options to consider, whereas a route with seven stops gives rise to 5,040 possible combinations. The number of possible combinations quickly grows into the millions as the number of stops increases. This phenomenon is known as factorial complexity, where the number of possible combinations increases much more rapidly than the number of stops. In fact, a route with 20 stops has 19 digits worth of possible combinations, while a route with 100 stops has a staggering 158 digits worth of possibilities. Below is an illustration of the planning complexity growing with a number of stops.

Interestingly enough, the human brain is capable of processing basic routing challenges quite quickly when a dispatcher can imagine geography, locations, and routes between destinations. The simple example below shows a few points and the optimal sequence to follow: 1-2-3-4.

As geography gets richer around natural barriers like rivers or mountains, both dispatchers and drivers may easily become puzzled while defining the optimal sequence to follow. For example, a river between points 1-3 and 2-4 with bridges available to cross. Having those natural boundaries, the optimal delivery sequence will be 1-3-2-4.

One-way streets and congested areas will introduce further limitations to the delivery logic and potential stop sequence. The example below demonstrates one-way bridges only letting traffic in one direction. The optimal sequence will be 1-3-4-2 in this case.

There are more potential aspects that will make sequencing tasks even harder. The example below shows preferred delivery windows, which may be introduced by distribution contracts. This changes the delivery sequence to 1-4-3-2. Many other aspects may give the same sort of challenges, such as onboard-time limits for fresh products, driving time limits.

It's important to keep in mind that even with only 4 delivery locations, the complexity of route planning can quickly become overwhelming. This becomes especially true when dealing with realistic scenarios where drivers are making 20-100 stops a day. It simply isn't possible for humans or machines to consider every possible combination and select the optimal delivery scenario. To solve this problem, modern scheduling systems use heuristics and algorithms to quickly generate an initial delivery plan and then improve it over time with human input.
It's also important to note that the examples given so far have only looked at a single delivery route in isolation, while the average delivery fleet may have as many as 8 vehicles to manage. In reality, the first step in the sequencing process is breaking down the full list of orders between vehicles.

Routing and Scheduling solutions are a class of software systems available in the form of client-server or SaaS applications in your browser. These solutions specialise in cracking delivery complexity for small and large fleets, and can typically save between 10% to 30% of mileage while optimising delivery routes.

Their success can be attributed to several factors:

They have knowledge of all potential distances between planned points. Even before initiating the planning algorithm, they calculate all potential route combinations between points. For example, if you plan for 100 drop operations, they will start from 10,000 potential routes to calculate mileage and driving time between them.
They have detailed road networks loaded in their memory to navigate from A to B using the most efficient routes while respecting one-way streets and allowed roads by vehicle types.
They have large hardware capacity running thousands of streams of calculations in parallel.
They use high-performance heuristics to look for optimal solutions using a powerful mathematical and statistical toolset.
They provide graphical planning interfaces that combine map and schedule views, which give dispatchers a visible and powerful way of amending the routes.

The implementation of route planning for a delivery fleet can significantly contribute to the sustainability of the planet. On average, each vehicle produces 28 kg of CO2 daily, based on a distance of 100 miles in the UK. For the typical SMB fleet of eight vehicles, a routing and scheduling solution could save up to 14 tons per year, with an average improvement in mileage of 23.2%. According to a study by the Aberdeen Group, companies that implemented routing and scheduling solutions saw an average reduction in travel time of 28.5%, a decrease in fuel usage of 23.2%, and an increase in workforce productivity of 15.2%.

Delivery efficiency

While route optimisation has a direct impact on the mileage of commercial fleets and environmental impact in general, everyday delivery efficiency would be another factor impacting the environment too.

Coming back to the migration from B2B to B2C delivery model, needless to say delivery fleets were not prepared for a new reality. As larger vehicles usually delivering to 10-20 business locations had to explore new residential areas making 30-70 multi-stop drops a day.
Key factors which made the new delivery job extremely complex were:

Missing driver’s knowledge of the delivery area
Unknown geography of residential districts including one way and narrow streets, traffic heavy junctions
Generic postcodes not giving accurate delivery locations
Dynamic nature of B2C delivery with order volumes changing on a daily basis.

The factors above put drivers in a stressful mindset, causing aggressive driving, driving at higher speeds and excessive idling. According to the US Department of Energy, aggressive driving (including speeding, rapid acceleration, and hard braking) can lower gas mileage by up to 33% on the highway and 5% in the city, resulting in increased CO2 emissions. The UK Department for Transport estimates that idling can use up to 2.6 litres of fuel per hour and emit 6.3 kg of CO2, which can add up over the course of a day or a week of delivery driving.

A delivery management solution can help overcome the key challenges associated with B2C deliveries, such as the driver's lack of knowledge of the delivery area and the dynamic nature of B2C delivery orders. By providing real-time, turn-by-turn navigation, the delivery management solution can ensure drivers are taking the most efficient route to each delivery location, even in unfamiliar residential areas. This can reduce the likelihood of missed deliveries, improve customer satisfaction, and minimise the environmental impact of driving.

Moreover, a delivery management solution can provide accurate delivery locations by leveraging more precise address information and geolocation data. This can help ensure that drivers are able to find each delivery location quickly and efficiently, reducing the amount of time they spend driving around looking for the right address. The system can also dynamically adjust delivery schedules based on changes in order volumes or traffic conditions, helping drivers to complete their deliveries on time and reducing the amount of time they spend idling and emitting CO2.

In conclusion, the implementation of route planning and delivery management solutions can have a significant positive impact on the environment. Route optimization can reduce mileage by up to 30%, resulting in a reduction of up to 14 tons of CO2 emissions per year for a typical SMB fleet. Additionally, delivery management solutions can improve delivery efficiency, reducing the likelihood of missed deliveries, improving customer satisfaction, and minimising the environmental impact of driving. By adopting these solutions, businesses can not only improve their bottom line but also contribute to a more sustainable future.

Meet the author: Aleksei Badianov

Aleksei Badianov is an accomplished and respected professional in the field of Product Management, known for his impressive achievements and exceptional leadership. With a primary focus on developing delivery route optimization solutions, Aleksei has successfully served numerous delivery and distribution companies across 12 countries. Through his innovative approach, he has optimized an impressive daily volume of over 100,000 deliveries, resulting in significant environmental benefits by reducing CO2 emissions by a remarkable 1,631,320 kg annually.
Beyond the tangible results, Aleksei's impact goes far beyond numbers. He has played a pivotal role in driving positive transformations within leading companies and cutting-edge startups. His unwavering commitment to personal and professional growth has kept him at the forefront of the rapidly evolving Product Management landscape.Throughout his career, Aleksei has consistently showcased his exceptional leadership skills by effectively leading teams of various sizes, ranging from 10 to 50 members, in both dynamic startup and corporate environments. His deep understanding of Product Management principles, coupled with his visionary approach, has yielded remarkable achievements and successful outcomes.

Aleksei's contributions and expertise have garnered recognition from esteemed Product Management communities, prestigious educational institutions, and mentoring platforms. He has become a trusted advisor and mentor, generously sharing his knowledge and insights to guide and empower aspiring Product Managers on their own career journeys.

With an unrelenting drive for improvement and a passion for innovation, Aleksei Badianov remains a highly sought-after professional in the Product Management field. His exceptional contributions, groundbreaking solutions, and unwavering dedication to personal growth stand as a testament to his remarkable abilities and enduring commitment to making a positive impact within the industry.

TRIZ: The Problem-Solving Methodology for Product Managers

Aleksei Badianov — Fri, 12 May 2023 14:32:11 +0000

As a product manager with over a decade of experience, I'm always looking for new ways to enhance my skills and help other product managers advance in their careers. One area that many companies focus on during the job interview process is analytical and creative problem-solving. And as product managers, we encounter these types of challenges on a daily basis. It's important to stay sharp and continuously develop our problem-solving abilities. That's why I made it a habit to practice a logic puzzle every day. But I also wondered if there was a common approach to solving these puzzles. That's when I discovered TRIZ – the Theory of Inventive Problem Solving. In this article, I'll provide an overview of what TRIZ is and how it can be applied in product management. Then, I'll apply TRIZ principles to solve a series of logical puzzles, showcasing the power and effectiveness of this methodology. So whether you're an experienced product manager looking to enhance your skills or someone who enjoys a good brain teaser, read on to discover the power of TRIZ!

If you find TRIZ to be a useful tool for problem-solving and innovation, there are many resources available to help you delve deeper into the methodology. The TRIZ Journal, for example, offers enough information on TRIZ, including case studies, articles, and other resources. You can also find books and online courses that provide a more in-depth look at TRIZ and how it can be applied in different industries and contexts. So if you're interested in learning more, there are plenty of opportunities to expand your knowledge and apply TRIZ to your work.

INTRO

Are you tired of boring, technical articles on problem-solving methodologies that put you to sleep? Well, fear not! This article on TRIZ promises to be different. We'll take a fresh look at how TRIZ can help you solve problems and innovate like a pro. Developed by a scientist Genrich Altshuller who was probably a lot more fun than his job title suggests, TRIZ is a problem-solving methodology that can help individuals and organizations overcome obstacles, reduce waste, and generate innovative solutions. So sit back, relax, and get ready to learn about TRIZ in a way that won't make you want to hit snooze.

TRIZ is a powerful problem-solving methodology that was created by the Soviet engineer and scientist Genrich Altshuller. Altshuller was a patent clerk who, in the 1940s, began analyzing patents to identify patterns and principles that could help inventors overcome obstacles and generate innovative solutions to complex challenges. This led to the development of TRIZ, which has since been used by organizations worldwide to enhance their product development processes and boost their overall innovation capabilities.

TRIZ includes 40 principles that are designed to guide problem-solving and innovation. These principles are based on the analysis of thousands of patents and the identification of patterns and solutions that can be applied to a wide range of problems. The TRIZ principles cover a broad range of topics, including reducing harm or waste, increasing efficiency, improving reliability, and enhancing functionality. Examples of these principles include the use of segmentation, the integration of parts, the use of porous materials, and the use of energy transitions. By understanding and applying these principles, individuals and organizations can overcome obstacles and generate more innovative and effective solutions to complex problems, leading to increased productivity, reduced costs, and improved customer satisfaction.

By leveraging TRIZ principles and tools, product managers can better identify and resolve issues, optimize existing products, and develop new ones that meet customer needs more effectively. This article will explore how TRIZ can be applied in product management and discuss some of the key benefits that this approach can offer to companies looking to improve their product development processes. From analyzing customer feedback to optimizing production processes, TRIZ can provide a structured approach for product managers to identify and address challenges, resulting in more innovative and successful products.

TRIZ Fundamentals

It's important to note that this article does not aim to provide a comprehensive analysis of each TRIZ principle and its application in product management. Rather, our goal is to offer a brief introduction to TRIZ as a tool for solving innovation challenges in the product development process.

TRIZ principles can be broadly categorized into four groups, also known as the four TRIZ innovation principles. These categories are Resources, Contradictions, Ideality, and Evolution.

The first group, Resources, includes principles that focus on how to make the most effective use of available resources, such as time, money, and materials. Examples of principles in this category include "Merging," which involves combining multiple functions or processes to save resources, and "Local Quality," which involves improving the quality of a specific part of a product or process to avoid wasting resources.
The second group, Contradictions, includes principles that are designed to help resolve contradictions or trade-offs between different aspects of a problem, such as cost versus quality. Examples of principles in this category include "Separation," which involves separating conflicting elements of a problem, and "Uniformity," which involves making different parts of a product or process more uniform to avoid contradictions.
The third group, Ideality, includes principles that aim to optimize a product or system by increasing its functionality while minimizing any negative impact on the environment or other factors. Examples of principles in this category include "Self-Service," which involves designing a product or process to require minimal external help, and "Simplicity," which involves reducing the number of components or processes to minimize waste.
The fourth and final group, Evolution, includes principles that focus on how to anticipate and prepare for changes in the market, technology, and other external factors, and how to leverage these changes to create new opportunities for innovation. Examples of principles in this category include "Continuity of Useful Action," which involves designing a product or process to continue functioning even when conditions change, and "Rapid Improvement," which involves quickly adapting to changes in the market or technology to stay ahead of the competition.

40 TRIZ PRINCIPLES

Each of four category includes principles that are designed to help individuals and organizations solve complex problems more effectively.

Resources. The 15 principles in this category are: Segmentation, Taking out, Local quality, Asymmetry, Merging, Universality, Nesting, Anti-weight, Preliminary anti-action, Prior counteraction, Cushioning, Equipotentiality, The other way round, Spheroidality, Dynamics.

Contradictions. The 16 principles in this category are: Universality, Preliminary action, Nesting, "Beforehand cushioning", Equipotentiality, "The other way around", "Spheroidality", "Dynamics", Partial or excessive actions, Another dimension, Another environment, Mechanical vibration, Periodic action, Preliminary action, The transition to a micro-level, Flexible shells and thin films, Porous materials.

Ideality. The 6 principles in this category are: Universality, Nesting, Equipotentiality, The other way round, Spheroidality, Dynamics, Segmentation.

Evolution. The 3 principles in this category are: Transition to a micro-level, Another dimension, Another environment.

PUZZLES: Ready to Exercise Your Brain?

As a product manager, one of the core responsibilities is to solve problems and develop innovative solutions that meet customer needs while balancing the constraints of time, budget, and resources. To accomplish this, product managers must be skilled in both analytical and creative thinking. I’ll use commonly known puzzles to illustrate how TRIZ can be applied to these challenges to generate more effective and innovative solutions.

This classic puzzle about the fox, hen, and corn is a popular game that has been enjoyed for centuries, and it's also frequently used as an interview question to test problem-solving abilities.

A farmer needs to transport a fox, a hen, and a sack of corn across a river. However, his boat is only big enough to transport one item at a time. If he leaves the fox alone with the hen, the fox will eat the hen. If he leaves the hen alone with the corn, the hen will eat the corn. How can the farmer transport all three items across the river without any of them being eaten?

The TRIZ principle of "The other way round" could be applied to the Fox/Hen/Corn puzzle. This principle suggests turning a problem or challenge on its head and looking at it from a different perspective. In the case of the puzzle, one could apply this principle by considering the possibility of transporting the fox, hen, and corn in the opposite direction across the river. By reversing the direction of transport, the solution to the problem may become clearer or easier to implement.

Base on this principle the solution to the puzzle will be:

The farmer takes the hen across the river and leaves it on the other side.
The farmer goes back across the river and takes the corn with him, but before leaving it on the other side, he brings the hen back to the original side.
The farmer leaves the hen on the original side and takes the fox with him to the other side.
He leaves the fox on the other side with the corn.
The farmer goes back across the river and brings the hen to the other side, leaving the corn and the fox on the other side.
Finally, the farmer goes back across the river one more time to get the corn and bring it to the other side, completing the task without any of the items being eaten.

THE ELECTRICITY CHALLENGE: Solve the Puzzle and Light Up Your Mind!

You are in a room with three light switches on the wall. Each switch controls one of three lamps in the next room, but you cannot see into the next room. You are allowed to flip the switches however you like, but you can only enter the next room once. How can you determine which switch controls each lamp?

TRIZ principle Partial or Excessive Actions could be applied to the bulb puzzle. The principle suggests modifying an action by dividing it into parts or increasing its intensity beyond its necessary level. In the context of the bulb puzzle, one possible application of this principle could be to turn on the first lamp for an excessive amount of time. By doing so, the first lamp would become significantly warmer than the other two lamps, which would remain cold. This would create a clearer distinction between the lamps, making it easier to identify which one is which when entering the room.

Based on this TRIZ principle a possible solution to the puzzle of the three lamps:

Turn on the first lamp and leave it on for a few minutes.
Turn off the first lamp and immediately turn on the second lamp.
Enter the room and feel the lamps. The lamp that is off but still warm to the touch is the first lamp, the lamp that is on is the second lamp, and the lamp that is off and cold is the third lamp.

PILL POWER

The puzzle involves a blind man who finds himself stranded on an uninhabited island with four pills - two blue and two red. He needs to take one red and one blue pill to survive, but taking two of the same color will result in his death. With no one else on the island to assist him, the blind man must figure out a way to safely take the correct combination of pills.

The Partial or Excessive Action principle suggests taking only a portion of an object or system to solve the problem. In this case, the blind man can take only a portion of each pill (e.g., half of each blue pill and half of each red pill) to achieve the desired combination of one red and one blue pill.

The Merging principle can also be applied by combining the two halves of the pills to form the desired combination. By breaking down the pills into smaller parts and then merging them in the correct way, the blind man can successfully solve the problem.

The solution to this puzzle is to split each of the pills in half, creating four halves in total. Then, the blind man can take one half of the blue pill and one half of the red pill, thus achieving his goal of getting one of each color without overdosing on either.

SOLVING THE CAMELS PUZZLE

A sheikh tells his two sons to race their camels to a distant city to see who will inherit his fortune. The one whose camel is slower will win. The brothers, after wandering aimlessly for days, ask a wise man for advice. After hearing the advice, they jump on the camels and race as fast as they can to the finish line. What advice did the wise man give them?

The TRIZ principle "Merging" suggests combining or merging different elements or components to create a new solution. In the case of the camel puzzle, the two sons could merge their resources by swapping camels, allowing them to both have a chance at winning the race and achieving their objective of having the camel come last.

The solution is for the two sons to swap their camels and ride as fast as possible to the distant city. The winner of the race will be the son whose camel comes second, since the objective is to have the camel come last.

WATER GLASSES PUZZLE

There are six glasses in a row. The first three are full of water, and the next three are empty. By moving only one glass how can you make them alternate between full and empty?

The TRIZ principle "Separation in Time or Space" suggests separating an object in time or space to allow for a change to occur, rather than directly manipulating the object itself. In the case of the glass puzzle, pouring water from one glass to another is a way to separate the water from the original glass and allow for the desired configuration to be achieved.

The solution is to pour the water from the second glass into the fifth glass, making the arrangement full, empty, full, empty, full, empty.

ROPES AND LIGHTERS

You have two ropes and a lighter. Each rope takes exactly one hour to burn from one end to the other. However, the rope doesn't burn at a consistent rate, so cutting the rope in half might not be a reliable way to measure 30 minutes. Using only these two ropes and a lighter, how can you measure exactly 45 minutes?

The Segmentation principle can be applied to the burning ropes puzzle. By breaking the rope into smaller segments, you can measure smaller time intervals.

The solution to the puzzle involves lighting one rope at both ends and the other rope at one end only. The first rope will burn completely in 30 minutes (since it is burning twice as fast), while the second rope will be half burned at that point. Then, the remaining half of the second rope can be ignited from the other end and will burn completely in 15 more minutes. This way, we get a total burning time of 45 minutes.

CONCLUSION

As we wrap up this article on TRIZ and problem-solving, it's important to note that TRIZ is not a magic wand that can solve all problems with a flick of the wrist. If only it were that easy! While TRIZ is a powerful tool for problem-solving and innovation, it still requires hard work, creativity, and persistence. But with TRIZ, you'll have a better framework for tackling those entertaining problems and coming up with more effective solutions.

So whether you're an experienced product manager looking to enhance your skills or a curious mind in search of a good challenge, give TRIZ a try. It's like the Swiss Army Knife of problem-solving methodologies – versatile, powerful, and always ready to help you tackle any problem that comes your way. And who knows, you might just discover a hidden talent for solving puzzles and impressing your colleagues with your newfound TRIZ skills. Happy problem-solving!

Meet the author: Aleksei Badianov

Listicles - a simple technique for benchmarking your value proposition against alternatives

Aleksei Badianov — Fri, 24 Feb 2023 11:21:11 +0000

Once upon a time there was a Product Manager

I have a killer product idea! It will be the “most awesomest” product out there. It will have this cool feature and this shiny interface, and blockchain, and AI, and a gravity gun, and blackjack, and... you know.....

Should my team go build it? Or should I first test if someone cares enough to buy it?

The case for demand validation

Ok, the books say I must validate the demand first, so let's be the diligent Senior Product Manager we should be and test if people want it before we build it.

We need to make sure that:

there is a real pain in the market that my cool product is a solution to.
the market is aware of the pain and cares enough to solve it.
my cool product can solve that pain point.
my cool product is a substantially better solution than the alternatives in the eyes of the market
I can communicate the value of my cool product in such a way as to motivate the market to choose it over the alternatives
with my intended pricing it will be competitive and attractive

oof, that's a lot of work! I'd rather just go build it, but fine.

Techniques for demand validation

What options for demand testing techniques do I have? Well, there are qualitative and quantitative methods, and I need both.
Qualitative method is relatively straightforward - I just need to interview enough people from the target market to dig into their jobs to be done, struggles and needs. They can also tell me about the alternative solutions they currently use to have their jobs done.
Quantitative method, on the other hand, is a bit tricky. I can always do a landing page test, but there are a few challenges with it:

Driving traffic to such a page is difficult and expensive, especially in a competitive market.
It's as much of a test for my ad, as it is for my product, and getting the ad right, considering its small size, is quite difficult and is a distraction.
But most importantly - I won't know how my solution can be compared to the available alternatives. There is no benchmark for the results I will get.

I have an idea though...

The listicles (list articles)

You know these "Top N something something" kind of articles? Like:

5 best GPS vehicle trackers
The 14 hair growth products that actually work
Top 10 Best CRM Software Tools in 2023

They are often referred to as "listicles" - articles presented in the form of a list.
They basically help you bypass doing your own research of possible solutions to a problem and present you with the best solutions to choose from.
They also offer an objective impartial analysis, rather than giving you a sales pitch designed to overstate benefits and downplay the drawbacks.
I love them - they make picking a new phone, a movie to watch, an app to install much easier. I also use them at work all the time while looking for solutions to everyday challenges.
So what if we use one of them to benchmark our product against best available alternatives?

Plan of action

Ok, here's my plan:

I will research available alternatives to solve the pain point my cool product is designed to solve.
I will write a "Top 5 solutions to {pain X}" article, where my cool product is one of those solutions.
I will make sure my article is an objective impartial analysis. I will not overstate my cool product's benefits and downplay its drawbacks. I will also be equally honest about the alternatives.
I will compare all solutions to the benefits I think my audience cares about the most, like price, ease of use, feature set and documentation.
I will host this article on a neutral domain (not on my product's website), but make sure I have access to the analytics.
I will promote this article using ads. It should be way easier and cheaper than promoting a specific solution as people generally don't like being sold to.
I will track the visitor behaviour - what they do once on the page, which links they click, which text they select. I might even do screen recordings of their actions (like HotJar) to understand how they behave on the page.
I will test how easy it is to drive traffic to this page. If people don't click it, it might mean that the pain point I am solving is not that important to my target market.
I will test how popular my solution is compared to the alternatives.
I will iterate on my value proposition until I am happy with how often users pick my solution.

Disciplining your thinking

I already see benefits of disciplining my thinking just from writing this article.

It forces me to clearly understand the problem I am solving.
I have to research the best available solutions to the problem.
I have no choice other than to think about differentiation very early.

Possible outcomes

Once I have this listicle up and getting traffic, here are possible outcomes:

It is possible I won't be able to drive significant traffic to the page. Since my market is established and not super-niche, that must mean that the market doesn't care enough about the problem.
Most obviously, visitors can prefer the alternatives over mine. I pretty much expect that in the first iteration. From this point I can A/B test various versions of my product's value prop to get to a good performance.
I can get to a point where I have a good number of visitors and a good conversion rate for my cool product compared to the alternatives. This will give me confidence to build the product and expect demand for it.

Summary

I believe listicles have a huge potential for testing demand hypotheses.

Writing one can discipline your thinking, forcing you to think through the problem, alternative solutions and differentiation at an early stage.
They are easier to promote, making demand validation more accessible.
They validate the problem by showing you early if people don't care enough to read it.
They give you a benchmark of your value proposition against best alternatives.

Have you tried using listicles for your demand validation? If so - let us know in the comments how this worked for you.

Meet the author: Aleksei Badianov

Will transport planners lose their jobs as AI becomes smarter?

Aleksei Badianov — Thu, 23 Feb 2023 23:41:35 +0000

As a Product Manager who has worked on the development of delivery route optimisation software for 10+ years, I see that modern technologies can significantly improve the optimisation process and deliver better solutions. AI, machine learning, and other modern technologies have the potential to revolutionise the way delivery routes are optimised in the future.

With the increasing availability of data and the advancement of AI and machine learning algorithms, it is becoming possible to develop more sophisticated prediction models that can be integrated into optimisation algorithms to make more accurate and informed decisions about route planning and scheduling. Machine learning algorithms can be trained to predict customer demand based on historical sales data and other market trends, allowing businesses to optimise their delivery schedules and routes accordingly. AI can also be used to optimise delivery schedules based on customer preferences and other relevant factors.

Blockchain technology could be used to create a secure, decentralised database of information about deliveries, including information about the products being shipped, the route they are taking, and the status of the delivery. This could help increase transparency and accountability in the delivery process as well as reduce the risk of fraud and theft.

Internet of Things (IoT) devices, such as sensors and GPS trackers, may collect real-time data about delivery vehicles and their surroundings. This data could be analysed and used to optimise delivery routes in real time, as well as to track the location of deliveries and monitor the condition of the products being shipped.

The use of real-time data and sensors can allow for more dynamic and adaptive route planning. By constantly monitoring traffic, weather, and other factors, delivery companies can adjust their routes and schedules in real time to avoid delays and optimise delivery times.

Overall, AI, machine learning, and other modern technologies have the potential to significantly improve the efficiency, accuracy, and sustainability of delivery route optimisation. As these technologies continue to evolve and become more widely adopted, we can expect to see significant improvements in the delivery industry in the future.

However, the question remains: can AI do all the delivery route optimisation without human intervention now? Is it just a matter of time before AI fully substitutes a human in the delivery route optimisation process?

AI-based delivery route optimisation software can create optimal routes in seconds, a task that would take humans hours or even days to complete. Additionally, these systems can consider and process vast amounts of data and provide real-time updates on the best routes.

Despite these advantages, there are still limitations to the use of AI in the delivery route optimisation process. For example, AI may not have the contextual knowledge and intuition that humans possess. Humans can make decisions based on experience, personal preferences, or ethical considerations, which AI may not be able to replicate.

Moreover, AI-based delivery route optimisation systems require significant data input and maintenance to work effectively. The accuracy and reliability of these systems depend on the quality and quantity of data they receive. Human intervention may be necessary to ensure that the data provided to the system is correct and up-to-date.

Therefore, while AI can optimise delivery routes effectively, it is unlikely that it can fully substitute humans in the delivery route optimisation process. A combination of AI and human expertise is likely to produce the best results. As technology advances, AI may be able to handle more complex delivery route optimisation tasks, but for now, humans remain an essential part of the process.

Moreover, it is important to carefully consider the potential risks and challenges associated with modern technologies and to implement them in a way that is safe, ethical, and sustainable.

Is route optimisation a hard job?
Efficient delivery route planning can help logistics companies reduce transportation costs, minimise delivery times while maximising the number of deliveries, and improve customer satisfaction. The optimisation of delivery routes can lead to significant cost savings, improve resource utilisation, and enhance the overall performance of logistics operations. The optimisation process involves finding the most cost-effective and efficient routes for delivery drivers, taking into account various factors such as distance, time, traffic, and vehicle capacity. That is why the delivery routes optimisation problem is crucial in logistics. Therefore, logistics companies need to invest in delivery route optimisation solutions to improve their competitiveness and profitability.

The delivery route optimisation process is complicated due to the number of variables involved, the need to balance competing priorities, and the constantly changing nature of delivery operations. Effective route optimisation requires sophisticated software, careful planning, and ongoing monitoring and adjustment to ensure that routes remain efficient and cost-effective over time.

The rise of advanced technology has led to the development of Route Optimisation Software that can significantly improve the efficiency of the delivery process. However, this has raised questions about whether it is possible to fully automate the delivery process and whether a human is still a necessary part. This article examines the above theses in more detail. Let me begin with the keystone of delivery route optimisation.

Types of goods
Different companies have different requirements for their delivery processes, depending on the type of goods they are shipping, the industry they operate in, the markets they serve, and their customers' needs. For example:

● Perishable goods, such as fresh produce, dairy products, or pharmaceuticals, have very strict requirements for temperature control and delivery speed to ensure that the products are delivered fresh and in good condition. Specialised transportation and storage facilities are the ground rules for them.

● Companies that ship high-value goods, such as jewellery, cash, or art, need to take extra security measures. Delivery routes should minimise the risk of theft or damage.

● Fragile goods, such as electronics, glassware, or art need to be protected during transportation by specialised packaging materials. Shipping companies plan delivery routes avoiding bumpy roads or other conditions that could damage the goods.

● To deliver heavy or oversized goods, such as construction equipment, vehicles, or furniture, companies use specialised transportation methods and equipment, such as flatbed trucks or cranes. The delivery route should avoid low bridges or other obstacles that could prevent the goods from being damaged.

To optimise their delivery processes, companies need to take these factors into account and develop customised strategies that meet their specific requirements.

Delivery types
When it comes to delivery route optimisation, the specific type of delivery service being offered can have a significant impact on the approach that is used. A delivery driver may have to visit dozens of stops in a single day, each with its own set of constraints and requirements. Some deliveries have to be made during specific time windows, while others may require special handling. Several types of delivery services exist, each with its unique characteristics and challenges:

● Point-to-point delivery is the most basic type of delivery service, where a single item or a small package is transported from one location to another.

● Last-mile delivery involves transporting goods from a local hub to the final destination, which is often a residential address.

● Same-day delivery means that goods should be delivered on the same day they are ordered. This type of delivery is becoming increasingly popular in the e-commerce industry.

● Scheduled delivery involves delivering goods at a pre-scheduled time, such as regular delivery of goods to a retailer or a grocery store.

In summary, different types of delivery services have different characteristics and requirements, and the choice of optimisation algorithm and approach will depend on the specific context and constraints of the delivery operation.

Conditions and human factors
Another challenge is the constantly changing conditions that affect delivery routes, such as road construction, bad weather conditions, and traffic patterns. These conditions can make it difficult to predict how long it will take to travel from one stop to another, making it challenging to plan the most efficient delivery route.

When something goes wrong in the delivery process, the delivery route and schedule should be rearranged to ensure that deliveries are made as efficiently as possible.

One way to rearrange it is to use specialised software that can recalculate the route and schedule and determine the best alternative based on the new parameters. This may involve resequencing the delivery orders, adjusting the delivery windows, and assigning new drivers or vehicles to the affected routes.

One of the most unpredictable things for any planning is the human factor. The behaviour of the driver, such as speeding, taking breaks, or deviating from the route, can influence the schedule. Moreover, if a driver is experiencing a health issue, they may not be able to drive safely, which can delay or disrupt the delivery process and put the driver at risk of car accidents. Accidents also cause delays, damage to the goods being delivered, and even result in injuries to the driver. This leads to additional costs for the business, and can negatively impact customer satisfaction.

To mitigate these risks, businesses should have contingency plans in place in case of driver illness or accidents, such as backup drivers or alternate delivery routes. Nevertheless, the human factor is usually not taken into account as a formula or a coefficient when calculating delivery routes. While optimisation algorithms and route planning software can take into account factors such as traffic patterns and vehicle capacity, they cannot directly incorporate the human element such as driver preferences, needs, or limitations.

Tools and approaches
As you can see, one of the most significant challenges associated with optimising delivery routes is the sheer number of variables that need to be considered. Special methods, approaches, and sophisticated software systems help to solve this problem:

● Geographic Information Systems (GIS) is software that is used to map out delivery locations and visualise the optimal routes, taking into account factors such as traffic, road conditions, and delivery schedules. GIS can also be used to analyse data and optimise routes based on factors such as distance, time, and cost.

● Real-time traffic data are used to optimise delivery routes based on current traffic conditions, which can help minimise delivery times and reduce fuel consumption.

● Vehicle telematics provides real-time data on vehicle location, speed, and other factors that can be used to optimise delivery routes.

● Data on customer location, delivery preferences, and order history can be used to optimise delivery routes and improve customer satisfaction.

● Optimisation approaches, such as divide and conquer nearest neighbour, and branch and bound can be used to optimise delivery routes manually, although this approach can be time-consuming and may not be suitable for complex problems.

● Human Expertise can also play a role in optimising delivery routes. Experienced logistics professionals can use their knowledge and expertise to identify opportunities for route optimisation, such as consolidating deliveries or rerouting vehicles to avoid traffic congestion.

There is a golden rule for any logistics company, that successful optimisation of delivery routes can be achieved only as a combination of all those methods and approaches. By utilising these tools, businesses can improve the efficiency and cost-effectiveness of their delivery operations, while also enhancing customer satisfaction.

Route optimisation software and algorithms
Route Optimisation Software is designed specifically for optimising delivery routes and combining all the best practices. This software uses algorithms to calculate the most efficient delivery routes based on a range of factors, including the number of stops, vehicle capacity, and delivery priorities. These software solutions use algorithms that can consider all of the variables and constraints:

● Nearest Neighbour Algorithm selects the closest delivery point to the current location and repeats the process until all delivery points have been visited.

● Clarke-Wright Algorithm is a heuristic method that builds routes by combining two delivery routes that are close to each other.

● Genetic Algorithm uses a genetic approach to create optimised routes. It works by creating a population of potential solutions and using selection, mutation, and recombination techniques to evolve the population toward the best solution.

● Ant Colony Optimisation is based on the behaviour of ants, which lay down pheromone trails to communicate with other ants. In this algorithm, many artificial ants are used to build routes by laying down and following pheromone trails.

● Simulated Annealing is based on the process of annealing in metallurgy, where a metal is heated and slowly cooled to reduce defects. In this algorithm, a solution is heated (or perturbed) and then cooled (or optimised) to find the optimal solution.

Another challenge is to choose the most relevant algorithm. The Nearest Neighbour Algorithm is suitable for small-scale problems and can quickly generate solutions. Otherwise, Genetic Algorithm or Ant Colony Optimisation is better to use for larger and more complex problems. Speaking of perishable goods, Simulated Annealing requires more time to produce optimal solutions, but it is usually used for problems that require maintaining the temperature of the goods being transported. Meanwhile, Clarke-Wright Algorithm effectively solves problems with vehicle capacity constraints.

These algorithms and others can be used in combination and the optimal algorithm may vary depending on the specific needs of the business and the problem at hand. Nevertheless, a mathematically optimised delivery route is not always equal to business-optimised logistics. While mathematical optimisation techniques can be very effective at identifying the most efficient route based on mathematical criteria, they may not always take into account real-world factors that can impact the delivery process.

For example, a mathematically optimised route may not take into account the preferences or requirements of individual customers, which could impact their satisfaction with the delivery service. Additionally, a mathematically optimised route may not take into account the availability of drivers or vehicles or other practical constraints that can impact the delivery process.

In some cases, there may be no optimal mathematical solution for a delivery route optimisation problem. This can occur when the problem is highly complex or when multiple competing objectives need to be balanced, such as minimising delivery time, reducing costs, and maximising customer satisfaction. In these cases, it may be necessary to use heuristic or simulation-based approaches, which can provide approximate solutions that are good enough for practical purposes.

Heuristics is a class of problem-solving techniques that are practical and useful when the optimal solution is unknown or difficult to calculate. They work by providing approximate solutions to complex problems in a reasonable amount of time.

One of the main advantages of heuristics algorithms for delivery route optimisation is that they can quickly generate good-quality solutions that are close to the optimal solution. In many cases, heuristics algorithms can find solutions that are almost as good as the optimal solution in a fraction of the time required by more computationally expensive optimisation algorithms.

Another advantage of heuristics algorithms is that they can be designed to incorporate practical constraints and requirements that are specific to the delivery route optimisation problem being solved.

To illustrate how it all works in real life let’s take a look at Maxoptra. It is a cloud-based software solution that is designed to help businesses optimise their delivery routes and schedules. Maxoptra’s main features are automated scheduling, real-time tracking, and dynamic routing, which can help businesses improve their operational efficiency, reduce delivery times, and minimise costs.

Maxoptra utilises a range of algorithms and techniques to optimise delivery routes, including heuristic algorithms, genetic algorithms, and Monte Carlo simulations. The software takes into account a variety of factors when calculating routes, including traffic patterns, weather conditions, road closures, and vehicle capacity.

One of the unique features of Maxoptra is its ability to dynamically update routes in real time, based on changing conditions. For example, if a driver experiences a delay or traffic jam, the software can automatically reroute the driver to the most efficient path to their destination. Therefore, only algorithms, including heuristics algorithms, in combination with human knowledge and expertise can provide the most effective solutions for delivery route optimisation.

Meet the author: Aleksei Badianov