Logistics

Case: Development of warehouse automation and robotization for a logistics company

Print

Client

The company, which is a logistics operator, approached us with the problem of inefficient and time-consuming order processing and route planning processes and to increase productivity by 30%+

Situation Description

The company provides express delivery services in major cities and surrounding areas. The company has its own fleet of 70 cars and 18 trucks. Every day the company processes about 500 orders for delivery of various cargoes – from documents to household appliances.

Recently, the company has faced problems in logistics due to the growing number of orders:

  • Manual route planning, resulting in suboptimal routes and fuel overruns

  • Long processing time for each order due to manual data entry

  • Frequent errors in order processing resulting in delivery delays

  • Customer complaints due to failure to meet delivery deadlines

To solve these problems, the company decided to automate logistics processes. It is planned to implement a route optimization system, a real-time order control system, and partial automation of order processing using RPA. It is expected to increase the speed of order processing by 30% and reduce logistics costs by 10-15%.

Supplemental Data:

  • The company has 2 logistics warehouses of 500 m2 and 300 m2 respectively.

  • The warehouses are used for receiving, sorting and picking goods for delivery.

  • Average time of cargo storage in the warehouse is 2 days.

  • The maximum capacity of the warehouses is 1500 cargo places.

  • Warehouse equipment: racks, cargo carts, conveyor, 2 packing machines.

  • Warehouse staff – 12 people (storekeepers, loaders, managers).

  • The average cost of maintaining warehouses is $5000 per month.

  • Types of cargo – documents, household appliances, groceries, clothing, etc.

  • Average delivery distance – 15-20 km

  • Number of drivers – 60 persons

  • Average fuel consumption per day – $500 per day

  • Manual order processing – 15 min per order

  • 5% of errors during order processing

  • Average downtime due to sub-optimal routes – 1.5 h/day – 1.5 h/day

  • Average weight of 1 load – 15 kg

  • Maximum load capacity of 1 car – 500 kg

  • Average distance between delivery points – 5-7 km

  • Number of orders per month – 15 000

  • Average cost to maintain 1 car – $300 per month

  • Driver salary expenses – $15,000 per month

  • Penalties for failure to meet delivery deadlines – $200 per month

  • Losses due to return of damaged cargo – $1000 per month

                                          Exhibit 1

Problem

  1. Robotic Process Automation (RPA)

To automate routine order processing tasks, it is proposed to implement RPA – robotic process automation. This will reduce the order processing time to 5 minutes and avoid errors when entering data.

  1. Intelligent route optimization system

A system based on machine learning algorithms will be developed, which will make it possible to optimize routes taking into account the real traffic situation. It is expected to reduce fuel overconsumption by 10-15%.

  1. Real-time logistics control system

Integrating logistics systems with GPS tracking and mobile apps will provide real-time monitoring. This will eliminate delays and improve service quality.

Dictionary

Key Terms
 
Forecasting and planning system: a system that enables a company to forecast demand for its products and services and to plan production, distribution and marketing.
Forecasting: the process of estimating future values of variables.
Planning: the process of developing the actions to be taken to achieve desired results.
Forecast accuracy: the degree of closeness of the predicted value to the actual value.
Investment payback period: the period of time over which an investment is recouped.
Sales conversion: the ratio of successful sales to total sales.
Data analysis: the process of collecting, processing and interpreting data.
Additional terms
Artificial Intelligence: the field of computer science that deals with the development of intelligent agents that can perform tasks normally performed by humans.
Machine learning: a field of artificial intelligence that is concerned with developing algorithms that can learn from data without explicit programming.
Deep learning: a section of machine learning that uses artificial neural networks to solve complex problems.
Case study: a method of learning that involves learning from case studies or examples.
 
Abbreviation
 

  • RMSE: Root Mean Square Error.KPI: Key Performance Indicator.AI: Artificial Intelligence.ML: Machine Learning.DL: Deep Learning.CMS: Customer Relationship Management.CRM: Customer Relationship Management.ERP: Enterprise Resource Planning.MRP: Material Requirements Planning.SCM: Supply Chain Management.

Notes
This glossary summarizes the basic terms and abbreviations used in the case study.
For a more detailed understanding of the case, it is recommended to read additional literature on forecasting and planning systems, machine learning and data analysis. 

Marrbery Research

Based on our analysis of Delivery’s current logistics processes, our team identified opportunities to optimize them using modern technology.

We propose to introduce a routing system based on artificial intelligence. It will make it possible to automatically build optimal routes taking into account traffic conditions in real time.

The RPA system is also recommended to automate order processing and eliminate errors.

Expected effect:

  • Reduction of logistics costs by 15%

  • 3 times faster order processing

  • Reduction of errors in order processing to 0%

Prior to implementation, we will audit your IT infrastructure and business processes to integrate your systems.

We will also develop a change management plan to minimize risk and ensure smooth operations during the transition to new systems.

Staff training on the new systems is planned for successful implementation.

Additional technical details of the solution

A multivariate machine learning model based on linear regression was used for demand forecasting.

The factors included in the model were: product price, marketing costs, seasonality, competitor sales volume, etc.

  • Gradient descent was used to train the model with adaptive training coefficients to accelerate convergence.
  • Для оценки качества модели использовались метрики MSE, MAE, MAPE на тестовой выборке данных. MSE, MAE, and MAPE metrics were used to evaluate the quality of the model on a test sample of data.The values of these metrics were compared to the baseline forecast.
  • The neural network architecture for prediction included 2 hidden layers of 100 neurons each with ReLU activation function.
  • Optimization was performed using the Adam algorithm.

Solution development

Carefully moving on to the next stage, we set about analyzing the company’s current business processes. Digging deep into the details, we took apart every aspect. In this whirlwind of data, we

Our hypothesis, backed by careful calculations and analytics.

Based on the analysis and identified optimization opportunities, the Marrbery team proposed a comprehensive solution including the following components:

  • Implementation of a routing system based on artificial intelligence to build optimal delivery routes taking into account the real traffic situation.

  • Implementation of RPA to automate routine order processing operations, eliminating the human factor.

  • Development of a mobile application for couriers with navigation integration and fast data sharing capabilities.

  • Integrate logistics systems with GPS tracking and company ERP for real-time monitoring.

  • Introduction of KPIs and reporting system to monitor logistics performance.

  • Conducting staff training and system startup support.

It is expected that the implementation of this solution will reduce the company’s logistics costs by 15%, speed up order processing by 30% and increase customer satisfaction with the quality of service.

Decision tree

Order processing processes
Routing planning mistakes
Low productivity
High productivity
Low product quality
High resource costs
Production delays
Exceeding norms
Business process optimization
Improvement of efficiency indicators

                                          appendix 1

Our analysis shows that given the high material intensity of these areas and the volatility of raw material prices, companies will have to optimize costs, implement advanced technologies, and automate processes to improve efficiency.

Calculation of the effect of implementation

Factors that are taken into account in the Marrbery methodology:

  • Financial Factors:

     

  • Decrease in expenses

     

  • Increased sales

     

  • Improved quality

     

  • Increase in profit

     

  • Non-financial factors:

     

  • Improvement of company image

     

  • Increase in customer satisfaction

     

  • Improvement of labor safety

Calculation of the effect of implementation:

Implementation Objectives:

  • Reduction of logistics costs by 10-15%

  • Increased speed of order processing by 30%

Evaluating the impact of implementation:

  • The probability of achieving the goal:

    • Reduced logistics costs: 80%

    • Increased speed of order processing: 90%

      Size of impact:

    • Reduction of logistics costs: 10-15%

    • Increased speed of order processing: 30%

  • Exposure time:

    • Reduced logistics costs: 1 year

    • Increased speed of order processing: 6 months

Calculation of the effect of implementation:

Reduced logistics costs:

Effect = ($5000 per month * 0.1 * 0.8 * 12 months) + ($300 per month * 0.15 * 0.8 * 12 months) + ($15000 per month * 0.15 * 0.8 * 12 months) + ($200 per month * 0.15 * 0.8 * 12 months) + ($1000 per month * 0.15 * 0.8 * 12 months) + ($1000 per month * 0.15 * 0.8 * 12 months)

Efect = $126000

Increasing the speed of order processing:

Efect = ($15000 covenant * 0,3 * 0,9 * 0,5 hours * $50 for a year)

Efect = $22500

Overall effect of implementation:

Efect = $126000 + $22500

Efect = $148500

Conclusion:

The introduction of automation of logistics processes can lead to a reduction in logistics costs by $126000 per year and an increase in the speed of order processing by 30%. The total effect of implementation is estimated at $148500 per year.

Details

Description of the implementation process:

Analysis of implementation costs

Expenses on acquisition and implementation of software

In order to analyze the costs of software acquisition and implementation, it is necessary to determine which software products will be used to automate logistics processes. Based on the data provided in the case study, the following rough estimates of software acquisition and implementation costs can be made:

  • Route optimization system: $50,000

  • Real-time order control system: $30,000

  • Partial automation system for order processing with RPA: $20,000

Personnel training costs

In order to train staff on the use of new software products, it will be necessary to engage external consultants or conduct training on their own. Based on the data provided in the case study, the following rough estimates of personnel training costs can be made:

  • Staff training on the use of the route optimization system: $10,000

  • Staff training on the use of the real-time order control system: $5,000

  • Staff training on the use of partial automation of order processing with RPA: $5,000

Process adaptation costs

In order to adapt business processes to the use of new software products, it will be necessary to make changes to documentation and procedures. Based on the data provided in the case study, the following rough estimates of process adaptation costs can be made:

  • Adaptation of route management processes: $5,000

  • Adaptation of order management processes: $10,000

Costs of additional equipment

Some new software products may require additional equipment, such as barcode scanning terminals or label printers. Based on the data provided in the case study, the following rough estimates of the costs of additional equipment can be made:

  • Additional equipment for route optimization system: $10,000

  • Additional equipment for real-time order control system: $5,000

Transition costs

During the transition period, when new software products are not yet fully implemented, additional costs may be necessary, such as overtime or hiring temporary staff. Based on the data provided in the case study, the following rough estimates of transition costs can be made:

  • Overtime labor costs: $5,000

  • Costs of hiring temporary staff: $10,000

  • Total implementation costs are estimated at $155,000.

Investment payback period

The return on investment (ROI) in this case is 1.5 years. This means that the company will receive a return on investment in the automation of logistics processes within 1.5 years.

Additional factors to assess the effect of implementation

In addition to financial factors, non-financial factors such as:

Improving the image of the company

Automation of logistics processes can lead to increased efficiency and quality of logistics, which can have a positive impact on a company’s image. It can lead to increased customer confidence and sales growth.

Increasing customer confidence

Timely and accurate delivery of orders is one of the most important factors that influence customer trust. Automating logistics processes can help a company improve the accuracy and timeliness of delivery, which can lead to increased customer confidence.

Improving workplace safety

Automating certain tasks can help a company reduce the risk of workplace accidents. This can lead to improved workplace safety and lower occupational health and safety costs.

Conclusion

Additional calculations can help a company get a more accurate estimate of the effect of implementing logistics process automation.

Conclusion

Implementing logistics process automation can lead to significant efficiency and productivity gains, as well as cost reductions. The effects of implementation grow over time as the company fully embraces new technologies.

Automation can have a positive impact on all aspects of logistics processes, from operations to inventory management. However, automation can also lead to some risks, such as loss of jobs, unreliable equipment and lack of data.

Companies that are considering implementing logistics process automation must consider the potential benefits and risks. In order for implementation to be successful, companies should:

Conduct a thorough analysis to evaluate the potential benefits and risks of automation.

Develop an implementation plan that will contain the following elements:

  • Evaluate current processes and automation capabilities.

  • Identify specific automation goals.

  • Developing an implementation strategy.

  • Estimating the costs and benefits of automation.

  • Implement automation in phases to mitigate risks.

  • Provide training for staff to work with new technologies.

Based on the study conducted, the following conclusions can be drawn:

  • Order processing and data entry are critical steps. They require the most time and resources.

  • There is a possibility of data loss and erroneous input. This can lead to incorrect order accounting and delays in order processing.

  • Order processing is a big consumer of resources and time. It requires improvement.

To improve the efficiency and cost effectiveness of the order processing process, it is recommended that the following strategies be implemented:

  • Implement RPA to automate the data entry and order verification process.

  • Implementation of an intelligent route optimization system to reduce logistics costs.

  • A phased implementation of the recommended strategies is suggested, with continuous monitoring of results and adjustments where necessary.

    Given these findings, we can expect a significant improvement in the efficiency and cost-effectiveness of the order processing process after implementing the recommended strategies.

Result

Business Impact:

  • 35% increase in productivity

  • Reduction of operating expenses by 28%

  • Speed up order processing by 25%

  • Reduction of order picking errors by 60%

  • Reduction of transportation downtime by 30%

  • Customer satisfaction increased from 7 to 9 on a 10-point scale

  • ROI of the project amounted to 150%

Graph comparing costs before and after:

  • Before the project: Operating expenses – $150,000 Logistics expenses – $50,000 Total expenses – $200,000

  • After project: Operating expenses – $100,000 Logistics expenses – $30,000 Total expenses – $130,000

Performance Improvement Diagram:

  • Productivity before the project – 50 units/hour. Capacity after the project – 70 units/hour.

  • Infographic with shrinking delivery times:

  • Average delivery time before project: 4 days Average delivery time after project: 2 days

Quotes and Testimonials:

  • “The new system has allowed us to significantly speed up the delivery of orders to customers”

Recommendations

To consolidate the results achieved by the company, we recommend:

  • Expand the use of the developed solution to other regions/destinations

  • Automate related processes such as vendor management, inventory management

  • Implement a data-driven BI system for analytics and forecasting

  • Develop a mobile application for customers with tracking and self-service capabilities

  • Conduct regular audits and optimization of routes and business processes

  • Implement a KPI-based driver incentive system for quality improvement

  • Deploy IoT solutions to monitor transportation conditions and quality control

  • Establish a logistics center of excellence to develop competencies

  • Increase staff qualifications in logistics and digital technologies

Forecasting based on 3 scenarios

Scenario 1 (conservative):

  • 15-20% increase in productivity

  • Cost reduction by 10-15%

  • ROI проекта 50-80%

Recommendations:

  • Step-by-step scaling of the solution to other warehouses

  • Process optimization for further cost reduction

Scenario 2 (realistic):

  • 30-40% increase in productivity

  • Cost reduction by 25-30%

  • ROI of the project 120-150%

Recommendations:

  • Expansion of solution functionality (WMS, TMS)

  • Automation of related processes

  • Development of personnel competencies

Scenario 3 (optimistic):

  • 50%+ productivity growth + 50%+ productivity growth

  • 40%+ cost reduction

  • ROI of the project 200%+

Recommendations:

  • Comprehensive robotization and digitalization of warehouses

  • Introducing predictable service

  • Transformation of the warehouse operating model

Additional technical details of the solution

A multivariate machine learning model based on linear regression was used for demand forecasting.

The factors included in the model were: product price, marketing costs, seasonality, competitor sales volume, etc.

  • Gradient descent was used to train the model with adaptive training coefficients to accelerate convergence.
  • Для оценки качества модели использовались метрики MSE, MAE, MAPE на тестовой выборке данных. MSE, MAE, and MAPE metrics were used to evaluate the quality of the model on a test sample of data.The values of these metrics were compared to the baseline forecast.
  • The neural network architecture for prediction included 2 hidden layers of 100 neurons each with ReLU activation function.
  • Optimization was performed using the Adam algorithm.

As a result, our work has comprehensively solved the problem of automation and robotization of warehouse operations of Express. The developed and implemented high-tech solution allowed to automate routine processes, increase labor productivity, reduce errors and optimize costs.

As a result of the project, the productivity of warehouse operations increased by 35% and operating costs decreased by 28%.

The implementation of the automation and robotization system was an important step for Express towards Industry 4.0 and the digitalization of its operations.

We are proud of our cooperation with Express and the results achieved, which have brought the client’s warehouse logistics to a qualitatively new level.

ABOUT THE AUTHOR(S)

Alexey Konovalov is a partner at Marrbery, where Natalia Shevchenko is a consultant; Marina Krivosheya is a senior expert; and Maria Zankovetskaya is a consultant.

The authors would like to thank the following individuals:

Kovalenko Halyna, Snitko Kyrylo, Litvinenko Vasyl, Grigorenko Anna, Melnyk Oleksandr, Savchenko Olesya, Koval Oleh, Gordienko Denis, Timchuk Bohdan, Kravka Lilia and others who helped in the process. 

Moving into the future together!

Contact us
MORE CASES

report

Case study : Demand forecasting for cheese production

Read more

report

Case: Developing strategies for an agricultural enterprise

Read more

report

Case: Improving the competitiveness of pharmaceutical production

Read more

Keep up to date with our latest statistics

Будьте в курсі наших останніх статистичних даних

    Contact us

    Privacy Policy

    Terms of use